]>
Commit | Line | Data |
---|---|---|
d77c3dfd | 1 | #!/usr/bin/env python |
dcdb292f | 2 | # coding: utf-8 |
d77c3dfd | 3 | |
ecc0c5ee PH |
4 | from __future__ import unicode_literals |
5 | ||
1e399778 | 6 | import base64 |
5bc880b9 | 7 | import binascii |
912b38b4 | 8 | import calendar |
676eb3f2 | 9 | import codecs |
62e609ab | 10 | import contextlib |
e3946f98 | 11 | import ctypes |
c496ca96 PH |
12 | import datetime |
13 | import email.utils | |
0c265486 | 14 | import email.header |
f45c185f | 15 | import errno |
be4a824d | 16 | import functools |
d77c3dfd | 17 | import gzip |
03f9daab | 18 | import io |
79a2e94e | 19 | import itertools |
f4bfd65f | 20 | import json |
d77c3dfd | 21 | import locale |
02dbf93f | 22 | import math |
347de493 | 23 | import operator |
d77c3dfd | 24 | import os |
c496ca96 | 25 | import platform |
773f291d | 26 | import random |
d77c3dfd | 27 | import re |
c496ca96 | 28 | import socket |
79a2e94e | 29 | import ssl |
1c088fa8 | 30 | import subprocess |
d77c3dfd | 31 | import sys |
181c8655 | 32 | import tempfile |
01951dda | 33 | import traceback |
bcf89ce6 | 34 | import xml.etree.ElementTree |
d77c3dfd | 35 | import zlib |
d77c3dfd | 36 | |
8c25f81b | 37 | from .compat import ( |
b4a3d461 | 38 | compat_HTMLParseError, |
8bb56eee | 39 | compat_HTMLParser, |
8f9312c3 | 40 | compat_basestring, |
8c25f81b | 41 | compat_chr, |
36e6f62c | 42 | compat_etree_fromstring, |
51098426 | 43 | compat_expanduser, |
8c25f81b | 44 | compat_html_entities, |
55b2f099 | 45 | compat_html_entities_html5, |
be4a824d | 46 | compat_http_client, |
c86b6142 | 47 | compat_kwargs, |
efa97bdc | 48 | compat_os_name, |
8c25f81b | 49 | compat_parse_qs, |
702ccf2d | 50 | compat_shlex_quote, |
be4a824d | 51 | compat_socket_create_connection, |
8c25f81b | 52 | compat_str, |
edaa23f8 | 53 | compat_struct_pack, |
d3f8e038 | 54 | compat_struct_unpack, |
8c25f81b PH |
55 | compat_urllib_error, |
56 | compat_urllib_parse, | |
15707c7e | 57 | compat_urllib_parse_urlencode, |
8c25f81b | 58 | compat_urllib_parse_urlparse, |
7581bfc9 | 59 | compat_urllib_parse_unquote_plus, |
8c25f81b PH |
60 | compat_urllib_request, |
61 | compat_urlparse, | |
810c10ba | 62 | compat_xpath, |
8c25f81b | 63 | ) |
4644ac55 | 64 | |
71aff188 YCH |
65 | from .socks import ( |
66 | ProxyType, | |
67 | sockssocket, | |
68 | ) | |
69 | ||
4644ac55 | 70 | |
51fb4995 YCH |
71 | def register_socks_protocols(): |
72 | # "Register" SOCKS protocols | |
d5ae6bb5 YCH |
73 | # In Python < 2.6.5, urlsplit() suffers from bug https://bugs.python.org/issue7904 |
74 | # URLs with protocols not in urlparse.uses_netloc are not handled correctly | |
51fb4995 YCH |
75 | for scheme in ('socks', 'socks4', 'socks4a', 'socks5'): |
76 | if scheme not in compat_urlparse.uses_netloc: | |
77 | compat_urlparse.uses_netloc.append(scheme) | |
78 | ||
79 | ||
468e2e92 FV |
80 | # This is not clearly defined otherwise |
81 | compiled_regex_type = type(re.compile('')) | |
82 | ||
3e669f36 | 83 | std_headers = { |
15d10678 | 84 | 'User-Agent': 'Mozilla/5.0 (X11; Linux x86_64; rv:10.0) Gecko/20150101 Firefox/47.0 (Chrome)', |
59ae15a5 PH |
85 | 'Accept-Charset': 'ISO-8859-1,utf-8;q=0.7,*;q=0.7', |
86 | 'Accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8', | |
87 | 'Accept-Encoding': 'gzip, deflate', | |
88 | 'Accept-Language': 'en-us,en;q=0.5', | |
3e669f36 | 89 | } |
f427df17 | 90 | |
5f6a1245 | 91 | |
fb37eb25 S |
92 | USER_AGENTS = { |
93 | 'Safari': 'Mozilla/5.0 (X11; Linux x86_64; rv:10.0) AppleWebKit/533.20.25 (KHTML, like Gecko) Version/5.0.4 Safari/533.20.27', | |
94 | } | |
95 | ||
96 | ||
bf42a990 S |
97 | NO_DEFAULT = object() |
98 | ||
7105440c YCH |
99 | ENGLISH_MONTH_NAMES = [ |
100 | 'January', 'February', 'March', 'April', 'May', 'June', | |
101 | 'July', 'August', 'September', 'October', 'November', 'December'] | |
102 | ||
f6717dec S |
103 | MONTH_NAMES = { |
104 | 'en': ENGLISH_MONTH_NAMES, | |
105 | 'fr': [ | |
3e4185c3 S |
106 | 'janvier', 'février', 'mars', 'avril', 'mai', 'juin', |
107 | 'juillet', 'août', 'septembre', 'octobre', 'novembre', 'décembre'], | |
f6717dec | 108 | } |
a942d6cb | 109 | |
a7aaa398 S |
110 | KNOWN_EXTENSIONS = ( |
111 | 'mp4', 'm4a', 'm4p', 'm4b', 'm4r', 'm4v', 'aac', | |
112 | 'flv', 'f4v', 'f4a', 'f4b', | |
113 | 'webm', 'ogg', 'ogv', 'oga', 'ogx', 'spx', 'opus', | |
114 | 'mkv', 'mka', 'mk3d', | |
115 | 'avi', 'divx', | |
116 | 'mov', | |
117 | 'asf', 'wmv', 'wma', | |
118 | '3gp', '3g2', | |
119 | 'mp3', | |
120 | 'flac', | |
121 | 'ape', | |
122 | 'wav', | |
123 | 'f4f', 'f4m', 'm3u8', 'smil') | |
124 | ||
c587cbb7 | 125 | # needed for sanitizing filenames in restricted mode |
c8827027 | 126 | ACCENT_CHARS = dict(zip('ÂÃÄÀÁÅÆÇÈÉÊËÌÍÎÏÐÑÒÓÔÕÖŐØŒÙÚÛÜŰÝÞßàáâãäåæçèéêëìíîïðñòóôõöőøœùúûüűýþÿ', |
127 | itertools.chain('AAAAAA', ['AE'], 'CEEEEIIIIDNOOOOOOO', ['OE'], 'UUUUUYP', ['ss'], | |
128 | 'aaaaaa', ['ae'], 'ceeeeiiiionooooooo', ['oe'], 'uuuuuypy'))) | |
c587cbb7 | 129 | |
46f59e89 S |
130 | DATE_FORMATS = ( |
131 | '%d %B %Y', | |
132 | '%d %b %Y', | |
133 | '%B %d %Y', | |
cb655f34 S |
134 | '%B %dst %Y', |
135 | '%B %dnd %Y', | |
136 | '%B %dth %Y', | |
46f59e89 | 137 | '%b %d %Y', |
cb655f34 S |
138 | '%b %dst %Y', |
139 | '%b %dnd %Y', | |
140 | '%b %dth %Y', | |
46f59e89 S |
141 | '%b %dst %Y %I:%M', |
142 | '%b %dnd %Y %I:%M', | |
143 | '%b %dth %Y %I:%M', | |
144 | '%Y %m %d', | |
145 | '%Y-%m-%d', | |
146 | '%Y/%m/%d', | |
81c13222 | 147 | '%Y/%m/%d %H:%M', |
46f59e89 | 148 | '%Y/%m/%d %H:%M:%S', |
0c1c6f4b | 149 | '%Y-%m-%d %H:%M', |
46f59e89 S |
150 | '%Y-%m-%d %H:%M:%S', |
151 | '%Y-%m-%d %H:%M:%S.%f', | |
152 | '%d.%m.%Y %H:%M', | |
153 | '%d.%m.%Y %H.%M', | |
154 | '%Y-%m-%dT%H:%M:%SZ', | |
155 | '%Y-%m-%dT%H:%M:%S.%fZ', | |
156 | '%Y-%m-%dT%H:%M:%S.%f0Z', | |
157 | '%Y-%m-%dT%H:%M:%S', | |
158 | '%Y-%m-%dT%H:%M:%S.%f', | |
159 | '%Y-%m-%dT%H:%M', | |
c6eed6b8 S |
160 | '%b %d %Y at %H:%M', |
161 | '%b %d %Y at %H:%M:%S', | |
46f59e89 S |
162 | ) |
163 | ||
164 | DATE_FORMATS_DAY_FIRST = list(DATE_FORMATS) | |
165 | DATE_FORMATS_DAY_FIRST.extend([ | |
166 | '%d-%m-%Y', | |
167 | '%d.%m.%Y', | |
168 | '%d.%m.%y', | |
169 | '%d/%m/%Y', | |
170 | '%d/%m/%y', | |
171 | '%d/%m/%Y %H:%M:%S', | |
172 | ]) | |
173 | ||
174 | DATE_FORMATS_MONTH_FIRST = list(DATE_FORMATS) | |
175 | DATE_FORMATS_MONTH_FIRST.extend([ | |
176 | '%m-%d-%Y', | |
177 | '%m.%d.%Y', | |
178 | '%m/%d/%Y', | |
179 | '%m/%d/%y', | |
180 | '%m/%d/%Y %H:%M:%S', | |
181 | ]) | |
182 | ||
06b3fe29 S |
183 | PACKED_CODES_RE = r"}\('(.+)',(\d+),(\d+),'([^']+)'\.split\('\|'\)" |
184 | ||
7105440c | 185 | |
d77c3dfd | 186 | def preferredencoding(): |
59ae15a5 | 187 | """Get preferred encoding. |
d77c3dfd | 188 | |
59ae15a5 PH |
189 | Returns the best encoding scheme for the system, based on |
190 | locale.getpreferredencoding() and some further tweaks. | |
191 | """ | |
192 | try: | |
193 | pref = locale.getpreferredencoding() | |
28e614de | 194 | 'TEST'.encode(pref) |
70a1165b | 195 | except Exception: |
59ae15a5 | 196 | pref = 'UTF-8' |
bae611f2 | 197 | |
59ae15a5 | 198 | return pref |
d77c3dfd | 199 | |
f4bfd65f | 200 | |
181c8655 | 201 | def write_json_file(obj, fn): |
1394646a | 202 | """ Encode obj as JSON and write it to fn, atomically if possible """ |
181c8655 | 203 | |
92120217 | 204 | fn = encodeFilename(fn) |
61ee5aeb | 205 | if sys.version_info < (3, 0) and sys.platform != 'win32': |
ec5f6016 JMF |
206 | encoding = get_filesystem_encoding() |
207 | # os.path.basename returns a bytes object, but NamedTemporaryFile | |
208 | # will fail if the filename contains non ascii characters unless we | |
209 | # use a unicode object | |
210 | path_basename = lambda f: os.path.basename(fn).decode(encoding) | |
211 | # the same for os.path.dirname | |
212 | path_dirname = lambda f: os.path.dirname(fn).decode(encoding) | |
213 | else: | |
214 | path_basename = os.path.basename | |
215 | path_dirname = os.path.dirname | |
216 | ||
73159f99 S |
217 | args = { |
218 | 'suffix': '.tmp', | |
ec5f6016 JMF |
219 | 'prefix': path_basename(fn) + '.', |
220 | 'dir': path_dirname(fn), | |
73159f99 S |
221 | 'delete': False, |
222 | } | |
223 | ||
181c8655 PH |
224 | # In Python 2.x, json.dump expects a bytestream. |
225 | # In Python 3.x, it writes to a character stream | |
226 | if sys.version_info < (3, 0): | |
73159f99 | 227 | args['mode'] = 'wb' |
181c8655 | 228 | else: |
73159f99 S |
229 | args.update({ |
230 | 'mode': 'w', | |
231 | 'encoding': 'utf-8', | |
232 | }) | |
233 | ||
c86b6142 | 234 | tf = tempfile.NamedTemporaryFile(**compat_kwargs(args)) |
181c8655 PH |
235 | |
236 | try: | |
237 | with tf: | |
238 | json.dump(obj, tf) | |
1394646a IK |
239 | if sys.platform == 'win32': |
240 | # Need to remove existing file on Windows, else os.rename raises | |
241 | # WindowsError or FileExistsError. | |
242 | try: | |
243 | os.unlink(fn) | |
244 | except OSError: | |
245 | pass | |
181c8655 | 246 | os.rename(tf.name, fn) |
70a1165b | 247 | except Exception: |
181c8655 PH |
248 | try: |
249 | os.remove(tf.name) | |
250 | except OSError: | |
251 | pass | |
252 | raise | |
253 | ||
254 | ||
255 | if sys.version_info >= (2, 7): | |
ee114368 | 256 | def find_xpath_attr(node, xpath, key, val=None): |
59ae56fa | 257 | """ Find the xpath xpath[@key=val] """ |
5d2354f1 | 258 | assert re.match(r'^[a-zA-Z_-]+$', key) |
ee114368 | 259 | expr = xpath + ('[@%s]' % key if val is None else "[@%s='%s']" % (key, val)) |
59ae56fa PH |
260 | return node.find(expr) |
261 | else: | |
ee114368 | 262 | def find_xpath_attr(node, xpath, key, val=None): |
810c10ba | 263 | for f in node.findall(compat_xpath(xpath)): |
ee114368 S |
264 | if key not in f.attrib: |
265 | continue | |
266 | if val is None or f.attrib.get(key) == val: | |
59ae56fa PH |
267 | return f |
268 | return None | |
269 | ||
d7e66d39 JMF |
270 | # On python2.6 the xml.etree.ElementTree.Element methods don't support |
271 | # the namespace parameter | |
5f6a1245 JW |
272 | |
273 | ||
d7e66d39 JMF |
274 | def xpath_with_ns(path, ns_map): |
275 | components = [c.split(':') for c in path.split('/')] | |
276 | replaced = [] | |
277 | for c in components: | |
278 | if len(c) == 1: | |
279 | replaced.append(c[0]) | |
280 | else: | |
281 | ns, tag = c | |
282 | replaced.append('{%s}%s' % (ns_map[ns], tag)) | |
283 | return '/'.join(replaced) | |
284 | ||
d77c3dfd | 285 | |
a41fb80c | 286 | def xpath_element(node, xpath, name=None, fatal=False, default=NO_DEFAULT): |
578c0745 | 287 | def _find_xpath(xpath): |
810c10ba | 288 | return node.find(compat_xpath(xpath)) |
578c0745 S |
289 | |
290 | if isinstance(xpath, (str, compat_str)): | |
291 | n = _find_xpath(xpath) | |
292 | else: | |
293 | for xp in xpath: | |
294 | n = _find_xpath(xp) | |
295 | if n is not None: | |
296 | break | |
d74bebd5 | 297 | |
8e636da4 | 298 | if n is None: |
bf42a990 S |
299 | if default is not NO_DEFAULT: |
300 | return default | |
301 | elif fatal: | |
bf0ff932 PH |
302 | name = xpath if name is None else name |
303 | raise ExtractorError('Could not find XML element %s' % name) | |
304 | else: | |
305 | return None | |
a41fb80c S |
306 | return n |
307 | ||
308 | ||
309 | def xpath_text(node, xpath, name=None, fatal=False, default=NO_DEFAULT): | |
8e636da4 S |
310 | n = xpath_element(node, xpath, name, fatal=fatal, default=default) |
311 | if n is None or n == default: | |
312 | return n | |
313 | if n.text is None: | |
314 | if default is not NO_DEFAULT: | |
315 | return default | |
316 | elif fatal: | |
317 | name = xpath if name is None else name | |
318 | raise ExtractorError('Could not find XML element\'s text %s' % name) | |
319 | else: | |
320 | return None | |
321 | return n.text | |
a41fb80c S |
322 | |
323 | ||
324 | def xpath_attr(node, xpath, key, name=None, fatal=False, default=NO_DEFAULT): | |
325 | n = find_xpath_attr(node, xpath, key) | |
326 | if n is None: | |
327 | if default is not NO_DEFAULT: | |
328 | return default | |
329 | elif fatal: | |
330 | name = '%s[@%s]' % (xpath, key) if name is None else name | |
331 | raise ExtractorError('Could not find XML attribute %s' % name) | |
332 | else: | |
333 | return None | |
334 | return n.attrib[key] | |
bf0ff932 PH |
335 | |
336 | ||
9e6dd238 | 337 | def get_element_by_id(id, html): |
43e8fafd | 338 | """Return the content of the tag with the specified ID in the passed HTML document""" |
611c1dd9 | 339 | return get_element_by_attribute('id', id, html) |
43e8fafd | 340 | |
12ea2f30 | 341 | |
84c237fb | 342 | def get_element_by_class(class_name, html): |
2af12ad9 TC |
343 | """Return the content of the first tag with the specified class in the passed HTML document""" |
344 | retval = get_elements_by_class(class_name, html) | |
345 | return retval[0] if retval else None | |
346 | ||
347 | ||
348 | def get_element_by_attribute(attribute, value, html, escape_value=True): | |
349 | retval = get_elements_by_attribute(attribute, value, html, escape_value) | |
350 | return retval[0] if retval else None | |
351 | ||
352 | ||
353 | def get_elements_by_class(class_name, html): | |
354 | """Return the content of all tags with the specified class in the passed HTML document as a list""" | |
355 | return get_elements_by_attribute( | |
84c237fb YCH |
356 | 'class', r'[^\'"]*\b%s\b[^\'"]*' % re.escape(class_name), |
357 | html, escape_value=False) | |
358 | ||
359 | ||
2af12ad9 | 360 | def get_elements_by_attribute(attribute, value, html, escape_value=True): |
43e8fafd | 361 | """Return the content of the tag with the specified attribute in the passed HTML document""" |
9e6dd238 | 362 | |
84c237fb YCH |
363 | value = re.escape(value) if escape_value else value |
364 | ||
2af12ad9 TC |
365 | retlist = [] |
366 | for m in re.finditer(r'''(?xs) | |
38285056 | 367 | <([a-zA-Z0-9:._-]+) |
609ff8ca | 368 | (?:\s+[a-zA-Z0-9:._-]+(?:=[a-zA-Z0-9:._-]*|="[^"]*"|='[^']*'|))*? |
38285056 | 369 | \s+%s=['"]?%s['"]? |
609ff8ca | 370 | (?:\s+[a-zA-Z0-9:._-]+(?:=[a-zA-Z0-9:._-]*|="[^"]*"|='[^']*'|))*? |
38285056 PH |
371 | \s*> |
372 | (?P<content>.*?) | |
373 | </\1> | |
2af12ad9 TC |
374 | ''' % (re.escape(attribute), value), html): |
375 | res = m.group('content') | |
38285056 | 376 | |
2af12ad9 TC |
377 | if res.startswith('"') or res.startswith("'"): |
378 | res = res[1:-1] | |
38285056 | 379 | |
2af12ad9 | 380 | retlist.append(unescapeHTML(res)) |
a921f407 | 381 | |
2af12ad9 | 382 | return retlist |
a921f407 | 383 | |
c5229f39 | 384 | |
8bb56eee BF |
385 | class HTMLAttributeParser(compat_HTMLParser): |
386 | """Trivial HTML parser to gather the attributes for a single element""" | |
387 | def __init__(self): | |
c5229f39 | 388 | self.attrs = {} |
8bb56eee BF |
389 | compat_HTMLParser.__init__(self) |
390 | ||
391 | def handle_starttag(self, tag, attrs): | |
392 | self.attrs = dict(attrs) | |
393 | ||
c5229f39 | 394 | |
8bb56eee BF |
395 | def extract_attributes(html_element): |
396 | """Given a string for an HTML element such as | |
397 | <el | |
398 | a="foo" B="bar" c="&98;az" d=boz | |
399 | empty= noval entity="&" | |
400 | sq='"' dq="'" | |
401 | > | |
402 | Decode and return a dictionary of attributes. | |
403 | { | |
404 | 'a': 'foo', 'b': 'bar', c: 'baz', d: 'boz', | |
405 | 'empty': '', 'noval': None, 'entity': '&', | |
406 | 'sq': '"', 'dq': '\'' | |
407 | }. | |
408 | NB HTMLParser is stricter in Python 2.6 & 3.2 than in later versions, | |
409 | but the cases in the unit test will work for all of 2.6, 2.7, 3.2-3.5. | |
410 | """ | |
411 | parser = HTMLAttributeParser() | |
b4a3d461 S |
412 | try: |
413 | parser.feed(html_element) | |
414 | parser.close() | |
415 | # Older Python may throw HTMLParseError in case of malformed HTML | |
416 | except compat_HTMLParseError: | |
417 | pass | |
8bb56eee | 418 | return parser.attrs |
9e6dd238 | 419 | |
c5229f39 | 420 | |
9e6dd238 | 421 | def clean_html(html): |
59ae15a5 | 422 | """Clean an HTML snippet into a readable string""" |
dd622d7c PH |
423 | |
424 | if html is None: # Convenience for sanitizing descriptions etc. | |
425 | return html | |
426 | ||
59ae15a5 PH |
427 | # Newline vs <br /> |
428 | html = html.replace('\n', ' ') | |
edd9221c TF |
429 | html = re.sub(r'(?u)\s*<\s*br\s*/?\s*>\s*', '\n', html) |
430 | html = re.sub(r'(?u)<\s*/\s*p\s*>\s*<\s*p[^>]*>', '\n', html) | |
59ae15a5 PH |
431 | # Strip html tags |
432 | html = re.sub('<.*?>', '', html) | |
433 | # Replace html entities | |
434 | html = unescapeHTML(html) | |
7decf895 | 435 | return html.strip() |
9e6dd238 FV |
436 | |
437 | ||
d77c3dfd | 438 | def sanitize_open(filename, open_mode): |
59ae15a5 PH |
439 | """Try to open the given filename, and slightly tweak it if this fails. |
440 | ||
441 | Attempts to open the given filename. If this fails, it tries to change | |
442 | the filename slightly, step by step, until it's either able to open it | |
443 | or it fails and raises a final exception, like the standard open() | |
444 | function. | |
445 | ||
446 | It returns the tuple (stream, definitive_file_name). | |
447 | """ | |
448 | try: | |
28e614de | 449 | if filename == '-': |
59ae15a5 PH |
450 | if sys.platform == 'win32': |
451 | import msvcrt | |
452 | msvcrt.setmode(sys.stdout.fileno(), os.O_BINARY) | |
898280a0 | 453 | return (sys.stdout.buffer if hasattr(sys.stdout, 'buffer') else sys.stdout, filename) |
59ae15a5 PH |
454 | stream = open(encodeFilename(filename), open_mode) |
455 | return (stream, filename) | |
456 | except (IOError, OSError) as err: | |
f45c185f PH |
457 | if err.errno in (errno.EACCES,): |
458 | raise | |
59ae15a5 | 459 | |
f45c185f | 460 | # In case of error, try to remove win32 forbidden chars |
d55de57b | 461 | alt_filename = sanitize_path(filename) |
f45c185f PH |
462 | if alt_filename == filename: |
463 | raise | |
464 | else: | |
465 | # An exception here should be caught in the caller | |
d55de57b | 466 | stream = open(encodeFilename(alt_filename), open_mode) |
f45c185f | 467 | return (stream, alt_filename) |
d77c3dfd FV |
468 | |
469 | ||
470 | def timeconvert(timestr): | |
59ae15a5 PH |
471 | """Convert RFC 2822 defined time string into system timestamp""" |
472 | timestamp = None | |
473 | timetuple = email.utils.parsedate_tz(timestr) | |
474 | if timetuple is not None: | |
475 | timestamp = email.utils.mktime_tz(timetuple) | |
476 | return timestamp | |
1c469a94 | 477 | |
5f6a1245 | 478 | |
796173d0 | 479 | def sanitize_filename(s, restricted=False, is_id=False): |
59ae15a5 PH |
480 | """Sanitizes a string so it could be used as part of a filename. |
481 | If restricted is set, use a stricter subset of allowed characters. | |
158af524 S |
482 | Set is_id if this is not an arbitrary string, but an ID that should be kept |
483 | if possible. | |
59ae15a5 PH |
484 | """ |
485 | def replace_insane(char): | |
c587cbb7 AT |
486 | if restricted and char in ACCENT_CHARS: |
487 | return ACCENT_CHARS[char] | |
59ae15a5 PH |
488 | if char == '?' or ord(char) < 32 or ord(char) == 127: |
489 | return '' | |
490 | elif char == '"': | |
491 | return '' if restricted else '\'' | |
492 | elif char == ':': | |
493 | return '_-' if restricted else ' -' | |
494 | elif char in '\\/|*<>': | |
495 | return '_' | |
627dcfff | 496 | if restricted and (char in '!&\'()[]{}$;`^,#' or char.isspace()): |
59ae15a5 PH |
497 | return '_' |
498 | if restricted and ord(char) > 127: | |
499 | return '_' | |
500 | return char | |
501 | ||
2aeb06d6 PH |
502 | # Handle timestamps |
503 | s = re.sub(r'[0-9]+(?::[0-9]+)+', lambda m: m.group(0).replace(':', '_'), s) | |
28e614de | 504 | result = ''.join(map(replace_insane, s)) |
796173d0 PH |
505 | if not is_id: |
506 | while '__' in result: | |
507 | result = result.replace('__', '_') | |
508 | result = result.strip('_') | |
509 | # Common case of "Foreign band name - English song title" | |
510 | if restricted and result.startswith('-_'): | |
511 | result = result[2:] | |
5a42414b PH |
512 | if result.startswith('-'): |
513 | result = '_' + result[len('-'):] | |
a7440261 | 514 | result = result.lstrip('.') |
796173d0 PH |
515 | if not result: |
516 | result = '_' | |
59ae15a5 | 517 | return result |
d77c3dfd | 518 | |
5f6a1245 | 519 | |
a2aaf4db S |
520 | def sanitize_path(s): |
521 | """Sanitizes and normalizes path on Windows""" | |
522 | if sys.platform != 'win32': | |
523 | return s | |
be531ef1 S |
524 | drive_or_unc, _ = os.path.splitdrive(s) |
525 | if sys.version_info < (2, 7) and not drive_or_unc: | |
526 | drive_or_unc, _ = os.path.splitunc(s) | |
527 | norm_path = os.path.normpath(remove_start(s, drive_or_unc)).split(os.path.sep) | |
528 | if drive_or_unc: | |
a2aaf4db S |
529 | norm_path.pop(0) |
530 | sanitized_path = [ | |
ec85ded8 | 531 | path_part if path_part in ['.', '..'] else re.sub(r'(?:[/<>:"\|\\?\*]|[\s.]$)', '#', path_part) |
a2aaf4db | 532 | for path_part in norm_path] |
be531ef1 S |
533 | if drive_or_unc: |
534 | sanitized_path.insert(0, drive_or_unc + os.path.sep) | |
a2aaf4db S |
535 | return os.path.join(*sanitized_path) |
536 | ||
537 | ||
67dda517 S |
538 | # Prepend protocol-less URLs with `http:` scheme in order to mitigate the number of |
539 | # unwanted failures due to missing protocol | |
17bcc626 S |
540 | def sanitize_url(url): |
541 | return 'http:%s' % url if url.startswith('//') else url | |
542 | ||
543 | ||
67dda517 | 544 | def sanitized_Request(url, *args, **kwargs): |
17bcc626 | 545 | return compat_urllib_request.Request(sanitize_url(url), *args, **kwargs) |
67dda517 S |
546 | |
547 | ||
51098426 S |
548 | def expand_path(s): |
549 | """Expand shell variables and ~""" | |
550 | return os.path.expandvars(compat_expanduser(s)) | |
551 | ||
552 | ||
d77c3dfd | 553 | def orderedSet(iterable): |
59ae15a5 PH |
554 | """ Remove all duplicates from the input iterable """ |
555 | res = [] | |
556 | for el in iterable: | |
557 | if el not in res: | |
558 | res.append(el) | |
559 | return res | |
d77c3dfd | 560 | |
912b38b4 | 561 | |
55b2f099 | 562 | def _htmlentity_transform(entity_with_semicolon): |
4e408e47 | 563 | """Transforms an HTML entity to a character.""" |
55b2f099 YCH |
564 | entity = entity_with_semicolon[:-1] |
565 | ||
4e408e47 PH |
566 | # Known non-numeric HTML entity |
567 | if entity in compat_html_entities.name2codepoint: | |
568 | return compat_chr(compat_html_entities.name2codepoint[entity]) | |
569 | ||
55b2f099 YCH |
570 | # TODO: HTML5 allows entities without a semicolon. For example, |
571 | # 'Éric' should be decoded as 'Éric'. | |
572 | if entity_with_semicolon in compat_html_entities_html5: | |
573 | return compat_html_entities_html5[entity_with_semicolon] | |
574 | ||
91757b0f | 575 | mobj = re.match(r'#(x[0-9a-fA-F]+|[0-9]+)', entity) |
4e408e47 PH |
576 | if mobj is not None: |
577 | numstr = mobj.group(1) | |
28e614de | 578 | if numstr.startswith('x'): |
4e408e47 | 579 | base = 16 |
28e614de | 580 | numstr = '0%s' % numstr |
4e408e47 PH |
581 | else: |
582 | base = 10 | |
7aefc49c S |
583 | # See https://github.com/rg3/youtube-dl/issues/7518 |
584 | try: | |
585 | return compat_chr(int(numstr, base)) | |
586 | except ValueError: | |
587 | pass | |
4e408e47 PH |
588 | |
589 | # Unknown entity in name, return its literal representation | |
7a3f0c00 | 590 | return '&%s;' % entity |
4e408e47 PH |
591 | |
592 | ||
d77c3dfd | 593 | def unescapeHTML(s): |
912b38b4 PH |
594 | if s is None: |
595 | return None | |
596 | assert type(s) == compat_str | |
d77c3dfd | 597 | |
4e408e47 | 598 | return re.sub( |
95f3f7c2 | 599 | r'&([^&;]+;)', lambda m: _htmlentity_transform(m.group(1)), s) |
d77c3dfd | 600 | |
8bf48f23 | 601 | |
aa49acd1 S |
602 | def get_subprocess_encoding(): |
603 | if sys.platform == 'win32' and sys.getwindowsversion()[0] >= 5: | |
604 | # For subprocess calls, encode with locale encoding | |
605 | # Refer to http://stackoverflow.com/a/9951851/35070 | |
606 | encoding = preferredencoding() | |
607 | else: | |
608 | encoding = sys.getfilesystemencoding() | |
609 | if encoding is None: | |
610 | encoding = 'utf-8' | |
611 | return encoding | |
612 | ||
613 | ||
8bf48f23 | 614 | def encodeFilename(s, for_subprocess=False): |
59ae15a5 PH |
615 | """ |
616 | @param s The name of the file | |
617 | """ | |
d77c3dfd | 618 | |
8bf48f23 | 619 | assert type(s) == compat_str |
d77c3dfd | 620 | |
59ae15a5 PH |
621 | # Python 3 has a Unicode API |
622 | if sys.version_info >= (3, 0): | |
623 | return s | |
0f00efed | 624 | |
aa49acd1 S |
625 | # Pass '' directly to use Unicode APIs on Windows 2000 and up |
626 | # (Detecting Windows NT 4 is tricky because 'major >= 4' would | |
627 | # match Windows 9x series as well. Besides, NT 4 is obsolete.) | |
628 | if not for_subprocess and sys.platform == 'win32' and sys.getwindowsversion()[0] >= 5: | |
629 | return s | |
630 | ||
8ee239e9 YCH |
631 | # Jython assumes filenames are Unicode strings though reported as Python 2.x compatible |
632 | if sys.platform.startswith('java'): | |
633 | return s | |
634 | ||
aa49acd1 S |
635 | return s.encode(get_subprocess_encoding(), 'ignore') |
636 | ||
637 | ||
638 | def decodeFilename(b, for_subprocess=False): | |
639 | ||
640 | if sys.version_info >= (3, 0): | |
641 | return b | |
642 | ||
643 | if not isinstance(b, bytes): | |
644 | return b | |
645 | ||
646 | return b.decode(get_subprocess_encoding(), 'ignore') | |
8bf48f23 | 647 | |
f07b74fc PH |
648 | |
649 | def encodeArgument(s): | |
650 | if not isinstance(s, compat_str): | |
651 | # Legacy code that uses byte strings | |
652 | # Uncomment the following line after fixing all post processors | |
7af808a5 | 653 | # assert False, 'Internal error: %r should be of type %r, is %r' % (s, compat_str, type(s)) |
f07b74fc PH |
654 | s = s.decode('ascii') |
655 | return encodeFilename(s, True) | |
656 | ||
657 | ||
aa49acd1 S |
658 | def decodeArgument(b): |
659 | return decodeFilename(b, True) | |
660 | ||
661 | ||
8271226a PH |
662 | def decodeOption(optval): |
663 | if optval is None: | |
664 | return optval | |
665 | if isinstance(optval, bytes): | |
666 | optval = optval.decode(preferredencoding()) | |
667 | ||
668 | assert isinstance(optval, compat_str) | |
669 | return optval | |
1c256f70 | 670 | |
5f6a1245 | 671 | |
4539dd30 PH |
672 | def formatSeconds(secs): |
673 | if secs > 3600: | |
674 | return '%d:%02d:%02d' % (secs // 3600, (secs % 3600) // 60, secs % 60) | |
675 | elif secs > 60: | |
676 | return '%d:%02d' % (secs // 60, secs % 60) | |
677 | else: | |
678 | return '%d' % secs | |
679 | ||
a0ddb8a2 | 680 | |
be4a824d PH |
681 | def make_HTTPS_handler(params, **kwargs): |
682 | opts_no_check_certificate = params.get('nocheckcertificate', False) | |
0db261ba | 683 | if hasattr(ssl, 'create_default_context'): # Python >= 3.4 or 2.7.9 |
be5f2c19 | 684 | context = ssl.create_default_context(ssl.Purpose.SERVER_AUTH) |
0db261ba | 685 | if opts_no_check_certificate: |
be5f2c19 | 686 | context.check_hostname = False |
0db261ba | 687 | context.verify_mode = ssl.CERT_NONE |
a2366922 | 688 | try: |
be4a824d | 689 | return YoutubeDLHTTPSHandler(params, context=context, **kwargs) |
a2366922 PH |
690 | except TypeError: |
691 | # Python 2.7.8 | |
692 | # (create_default_context present but HTTPSHandler has no context=) | |
693 | pass | |
694 | ||
695 | if sys.version_info < (3, 2): | |
d7932313 | 696 | return YoutubeDLHTTPSHandler(params, **kwargs) |
aa37e3d4 | 697 | else: # Python < 3.4 |
d7932313 | 698 | context = ssl.SSLContext(ssl.PROTOCOL_TLSv1) |
ea6d901e | 699 | context.verify_mode = (ssl.CERT_NONE |
dca08720 | 700 | if opts_no_check_certificate |
ea6d901e | 701 | else ssl.CERT_REQUIRED) |
303b479e | 702 | context.set_default_verify_paths() |
be4a824d | 703 | return YoutubeDLHTTPSHandler(params, context=context, **kwargs) |
ea6d901e | 704 | |
732ea2f0 | 705 | |
08f2a92c JMF |
706 | def bug_reports_message(): |
707 | if ytdl_is_updateable(): | |
708 | update_cmd = 'type youtube-dl -U to update' | |
709 | else: | |
710 | update_cmd = 'see https://yt-dl.org/update on how to update' | |
711 | msg = '; please report this issue on https://yt-dl.org/bug .' | |
712 | msg += ' Make sure you are using the latest version; %s.' % update_cmd | |
713 | msg += ' Be sure to call youtube-dl with the --verbose flag and include its complete output.' | |
714 | return msg | |
715 | ||
716 | ||
bf5b9d85 PM |
717 | class YoutubeDLError(Exception): |
718 | """Base exception for YoutubeDL errors.""" | |
719 | pass | |
720 | ||
721 | ||
722 | class ExtractorError(YoutubeDLError): | |
1c256f70 | 723 | """Error during info extraction.""" |
5f6a1245 | 724 | |
d11271dd | 725 | def __init__(self, msg, tb=None, expected=False, cause=None, video_id=None): |
9a82b238 PH |
726 | """ tb, if given, is the original traceback (so that it can be printed out). |
727 | If expected is set, this is a normal error message and most likely not a bug in youtube-dl. | |
728 | """ | |
729 | ||
730 | if sys.exc_info()[0] in (compat_urllib_error.URLError, socket.timeout, UnavailableVideoError): | |
731 | expected = True | |
d11271dd PH |
732 | if video_id is not None: |
733 | msg = video_id + ': ' + msg | |
410f3e73 | 734 | if cause: |
28e614de | 735 | msg += ' (caused by %r)' % cause |
9a82b238 | 736 | if not expected: |
08f2a92c | 737 | msg += bug_reports_message() |
1c256f70 | 738 | super(ExtractorError, self).__init__(msg) |
d5979c5d | 739 | |
1c256f70 | 740 | self.traceback = tb |
8cc83b8d | 741 | self.exc_info = sys.exc_info() # preserve original exception |
2eabb802 | 742 | self.cause = cause |
d11271dd | 743 | self.video_id = video_id |
1c256f70 | 744 | |
01951dda PH |
745 | def format_traceback(self): |
746 | if self.traceback is None: | |
747 | return None | |
28e614de | 748 | return ''.join(traceback.format_tb(self.traceback)) |
01951dda | 749 | |
1c256f70 | 750 | |
416c7fcb PH |
751 | class UnsupportedError(ExtractorError): |
752 | def __init__(self, url): | |
753 | super(UnsupportedError, self).__init__( | |
754 | 'Unsupported URL: %s' % url, expected=True) | |
755 | self.url = url | |
756 | ||
757 | ||
55b3e45b JMF |
758 | class RegexNotFoundError(ExtractorError): |
759 | """Error when a regex didn't match""" | |
760 | pass | |
761 | ||
762 | ||
773f291d S |
763 | class GeoRestrictedError(ExtractorError): |
764 | """Geographic restriction Error exception. | |
765 | ||
766 | This exception may be thrown when a video is not available from your | |
767 | geographic location due to geographic restrictions imposed by a website. | |
768 | """ | |
769 | def __init__(self, msg, countries=None): | |
770 | super(GeoRestrictedError, self).__init__(msg, expected=True) | |
771 | self.msg = msg | |
772 | self.countries = countries | |
773 | ||
774 | ||
bf5b9d85 | 775 | class DownloadError(YoutubeDLError): |
59ae15a5 | 776 | """Download Error exception. |
d77c3dfd | 777 | |
59ae15a5 PH |
778 | This exception may be thrown by FileDownloader objects if they are not |
779 | configured to continue on errors. They will contain the appropriate | |
780 | error message. | |
781 | """ | |
5f6a1245 | 782 | |
8cc83b8d FV |
783 | def __init__(self, msg, exc_info=None): |
784 | """ exc_info, if given, is the original exception that caused the trouble (as returned by sys.exc_info()). """ | |
785 | super(DownloadError, self).__init__(msg) | |
786 | self.exc_info = exc_info | |
d77c3dfd FV |
787 | |
788 | ||
bf5b9d85 | 789 | class SameFileError(YoutubeDLError): |
59ae15a5 | 790 | """Same File exception. |
d77c3dfd | 791 | |
59ae15a5 PH |
792 | This exception will be thrown by FileDownloader objects if they detect |
793 | multiple files would have to be downloaded to the same file on disk. | |
794 | """ | |
795 | pass | |
d77c3dfd FV |
796 | |
797 | ||
bf5b9d85 | 798 | class PostProcessingError(YoutubeDLError): |
59ae15a5 | 799 | """Post Processing exception. |
d77c3dfd | 800 | |
59ae15a5 PH |
801 | This exception may be raised by PostProcessor's .run() method to |
802 | indicate an error in the postprocessing task. | |
803 | """ | |
5f6a1245 | 804 | |
7851b379 | 805 | def __init__(self, msg): |
bf5b9d85 | 806 | super(PostProcessingError, self).__init__(msg) |
7851b379 | 807 | self.msg = msg |
d77c3dfd | 808 | |
5f6a1245 | 809 | |
bf5b9d85 | 810 | class MaxDownloadsReached(YoutubeDLError): |
59ae15a5 PH |
811 | """ --max-downloads limit has been reached. """ |
812 | pass | |
d77c3dfd FV |
813 | |
814 | ||
bf5b9d85 | 815 | class UnavailableVideoError(YoutubeDLError): |
59ae15a5 | 816 | """Unavailable Format exception. |
d77c3dfd | 817 | |
59ae15a5 PH |
818 | This exception will be thrown when a video is requested |
819 | in a format that is not available for that video. | |
820 | """ | |
821 | pass | |
d77c3dfd FV |
822 | |
823 | ||
bf5b9d85 | 824 | class ContentTooShortError(YoutubeDLError): |
59ae15a5 | 825 | """Content Too Short exception. |
d77c3dfd | 826 | |
59ae15a5 PH |
827 | This exception may be raised by FileDownloader objects when a file they |
828 | download is too small for what the server announced first, indicating | |
829 | the connection was probably interrupted. | |
830 | """ | |
d77c3dfd | 831 | |
59ae15a5 | 832 | def __init__(self, downloaded, expected): |
bf5b9d85 PM |
833 | super(ContentTooShortError, self).__init__( |
834 | 'Downloaded {0} bytes, expected {1} bytes'.format(downloaded, expected) | |
835 | ) | |
2c7ed247 | 836 | # Both in bytes |
59ae15a5 PH |
837 | self.downloaded = downloaded |
838 | self.expected = expected | |
d77c3dfd | 839 | |
5f6a1245 | 840 | |
bf5b9d85 | 841 | class XAttrMetadataError(YoutubeDLError): |
efa97bdc YCH |
842 | def __init__(self, code=None, msg='Unknown error'): |
843 | super(XAttrMetadataError, self).__init__(msg) | |
844 | self.code = code | |
bd264412 | 845 | self.msg = msg |
efa97bdc YCH |
846 | |
847 | # Parsing code and msg | |
848 | if (self.code in (errno.ENOSPC, errno.EDQUOT) or | |
849 | 'No space left' in self.msg or 'Disk quota excedded' in self.msg): | |
850 | self.reason = 'NO_SPACE' | |
851 | elif self.code == errno.E2BIG or 'Argument list too long' in self.msg: | |
852 | self.reason = 'VALUE_TOO_LONG' | |
853 | else: | |
854 | self.reason = 'NOT_SUPPORTED' | |
855 | ||
856 | ||
bf5b9d85 | 857 | class XAttrUnavailableError(YoutubeDLError): |
efa97bdc YCH |
858 | pass |
859 | ||
860 | ||
c5a59d93 | 861 | def _create_http_connection(ydl_handler, http_class, is_https, *args, **kwargs): |
e5e78797 S |
862 | # Working around python 2 bug (see http://bugs.python.org/issue17849) by limiting |
863 | # expected HTTP responses to meet HTTP/1.0 or later (see also | |
864 | # https://github.com/rg3/youtube-dl/issues/6727) | |
865 | if sys.version_info < (3, 0): | |
5a1a2e94 | 866 | kwargs[b'strict'] = True |
be4a824d PH |
867 | hc = http_class(*args, **kwargs) |
868 | source_address = ydl_handler._params.get('source_address') | |
869 | if source_address is not None: | |
870 | sa = (source_address, 0) | |
871 | if hasattr(hc, 'source_address'): # Python 2.7+ | |
872 | hc.source_address = sa | |
873 | else: # Python 2.6 | |
874 | def _hc_connect(self, *args, **kwargs): | |
875 | sock = compat_socket_create_connection( | |
876 | (self.host, self.port), self.timeout, sa) | |
877 | if is_https: | |
d7932313 PH |
878 | self.sock = ssl.wrap_socket( |
879 | sock, self.key_file, self.cert_file, | |
880 | ssl_version=ssl.PROTOCOL_TLSv1) | |
be4a824d PH |
881 | else: |
882 | self.sock = sock | |
883 | hc.connect = functools.partial(_hc_connect, hc) | |
884 | ||
885 | return hc | |
886 | ||
887 | ||
87f0e62d | 888 | def handle_youtubedl_headers(headers): |
992fc9d6 YCH |
889 | filtered_headers = headers |
890 | ||
891 | if 'Youtubedl-no-compression' in filtered_headers: | |
892 | filtered_headers = dict((k, v) for k, v in filtered_headers.items() if k.lower() != 'accept-encoding') | |
87f0e62d | 893 | del filtered_headers['Youtubedl-no-compression'] |
87f0e62d | 894 | |
992fc9d6 | 895 | return filtered_headers |
87f0e62d YCH |
896 | |
897 | ||
acebc9cd | 898 | class YoutubeDLHandler(compat_urllib_request.HTTPHandler): |
59ae15a5 PH |
899 | """Handler for HTTP requests and responses. |
900 | ||
901 | This class, when installed with an OpenerDirector, automatically adds | |
902 | the standard headers to every HTTP request and handles gzipped and | |
903 | deflated responses from web servers. If compression is to be avoided in | |
904 | a particular request, the original request in the program code only has | |
0424ec30 | 905 | to include the HTTP header "Youtubedl-no-compression", which will be |
59ae15a5 PH |
906 | removed before making the real request. |
907 | ||
908 | Part of this code was copied from: | |
909 | ||
910 | http://techknack.net/python-urllib2-handlers/ | |
911 | ||
912 | Andrew Rowls, the author of that code, agreed to release it to the | |
913 | public domain. | |
914 | """ | |
915 | ||
be4a824d PH |
916 | def __init__(self, params, *args, **kwargs): |
917 | compat_urllib_request.HTTPHandler.__init__(self, *args, **kwargs) | |
918 | self._params = params | |
919 | ||
920 | def http_open(self, req): | |
71aff188 YCH |
921 | conn_class = compat_http_client.HTTPConnection |
922 | ||
923 | socks_proxy = req.headers.get('Ytdl-socks-proxy') | |
924 | if socks_proxy: | |
925 | conn_class = make_socks_conn_class(conn_class, socks_proxy) | |
926 | del req.headers['Ytdl-socks-proxy'] | |
927 | ||
be4a824d | 928 | return self.do_open(functools.partial( |
71aff188 | 929 | _create_http_connection, self, conn_class, False), |
be4a824d PH |
930 | req) |
931 | ||
59ae15a5 PH |
932 | @staticmethod |
933 | def deflate(data): | |
934 | try: | |
935 | return zlib.decompress(data, -zlib.MAX_WBITS) | |
936 | except zlib.error: | |
937 | return zlib.decompress(data) | |
938 | ||
acebc9cd | 939 | def http_request(self, req): |
51f267d9 S |
940 | # According to RFC 3986, URLs can not contain non-ASCII characters, however this is not |
941 | # always respected by websites, some tend to give out URLs with non percent-encoded | |
942 | # non-ASCII characters (see telemb.py, ard.py [#3412]) | |
943 | # urllib chokes on URLs with non-ASCII characters (see http://bugs.python.org/issue3991) | |
944 | # To work around aforementioned issue we will replace request's original URL with | |
945 | # percent-encoded one | |
946 | # Since redirects are also affected (e.g. http://www.southpark.de/alle-episoden/s18e09) | |
947 | # the code of this workaround has been moved here from YoutubeDL.urlopen() | |
948 | url = req.get_full_url() | |
949 | url_escaped = escape_url(url) | |
950 | ||
951 | # Substitute URL if any change after escaping | |
952 | if url != url_escaped: | |
15d260eb | 953 | req = update_Request(req, url=url_escaped) |
51f267d9 | 954 | |
33ac271b | 955 | for h, v in std_headers.items(): |
3d5f7a39 JK |
956 | # Capitalize is needed because of Python bug 2275: http://bugs.python.org/issue2275 |
957 | # The dict keys are capitalized because of this bug by urllib | |
958 | if h.capitalize() not in req.headers: | |
33ac271b | 959 | req.add_header(h, v) |
87f0e62d YCH |
960 | |
961 | req.headers = handle_youtubedl_headers(req.headers) | |
989b4b2b PH |
962 | |
963 | if sys.version_info < (2, 7) and '#' in req.get_full_url(): | |
964 | # Python 2.6 is brain-dead when it comes to fragments | |
965 | req._Request__original = req._Request__original.partition('#')[0] | |
966 | req._Request__r_type = req._Request__r_type.partition('#')[0] | |
967 | ||
59ae15a5 PH |
968 | return req |
969 | ||
acebc9cd | 970 | def http_response(self, req, resp): |
59ae15a5 PH |
971 | old_resp = resp |
972 | # gzip | |
973 | if resp.headers.get('Content-encoding', '') == 'gzip': | |
aa3e9507 PH |
974 | content = resp.read() |
975 | gz = gzip.GzipFile(fileobj=io.BytesIO(content), mode='rb') | |
976 | try: | |
977 | uncompressed = io.BytesIO(gz.read()) | |
978 | except IOError as original_ioerror: | |
979 | # There may be junk add the end of the file | |
980 | # See http://stackoverflow.com/q/4928560/35070 for details | |
981 | for i in range(1, 1024): | |
982 | try: | |
983 | gz = gzip.GzipFile(fileobj=io.BytesIO(content[:-i]), mode='rb') | |
984 | uncompressed = io.BytesIO(gz.read()) | |
985 | except IOError: | |
986 | continue | |
987 | break | |
988 | else: | |
989 | raise original_ioerror | |
b407d853 | 990 | resp = compat_urllib_request.addinfourl(uncompressed, old_resp.headers, old_resp.url, old_resp.code) |
59ae15a5 | 991 | resp.msg = old_resp.msg |
c047270c | 992 | del resp.headers['Content-encoding'] |
59ae15a5 PH |
993 | # deflate |
994 | if resp.headers.get('Content-encoding', '') == 'deflate': | |
995 | gz = io.BytesIO(self.deflate(resp.read())) | |
b407d853 | 996 | resp = compat_urllib_request.addinfourl(gz, old_resp.headers, old_resp.url, old_resp.code) |
59ae15a5 | 997 | resp.msg = old_resp.msg |
c047270c | 998 | del resp.headers['Content-encoding'] |
ad729172 S |
999 | # Percent-encode redirect URL of Location HTTP header to satisfy RFC 3986 (see |
1000 | # https://github.com/rg3/youtube-dl/issues/6457). | |
5a4d9ddb S |
1001 | if 300 <= resp.code < 400: |
1002 | location = resp.headers.get('Location') | |
1003 | if location: | |
1004 | # As of RFC 2616 default charset is iso-8859-1 that is respected by python 3 | |
1005 | if sys.version_info >= (3, 0): | |
1006 | location = location.encode('iso-8859-1').decode('utf-8') | |
0ea59007 YCH |
1007 | else: |
1008 | location = location.decode('utf-8') | |
5a4d9ddb S |
1009 | location_escaped = escape_url(location) |
1010 | if location != location_escaped: | |
1011 | del resp.headers['Location'] | |
9a4aec8b YCH |
1012 | if sys.version_info < (3, 0): |
1013 | location_escaped = location_escaped.encode('utf-8') | |
5a4d9ddb | 1014 | resp.headers['Location'] = location_escaped |
59ae15a5 | 1015 | return resp |
0f8d03f8 | 1016 | |
acebc9cd PH |
1017 | https_request = http_request |
1018 | https_response = http_response | |
bf50b038 | 1019 | |
5de90176 | 1020 | |
71aff188 YCH |
1021 | def make_socks_conn_class(base_class, socks_proxy): |
1022 | assert issubclass(base_class, ( | |
1023 | compat_http_client.HTTPConnection, compat_http_client.HTTPSConnection)) | |
1024 | ||
1025 | url_components = compat_urlparse.urlparse(socks_proxy) | |
1026 | if url_components.scheme.lower() == 'socks5': | |
1027 | socks_type = ProxyType.SOCKS5 | |
1028 | elif url_components.scheme.lower() in ('socks', 'socks4'): | |
1029 | socks_type = ProxyType.SOCKS4 | |
51fb4995 YCH |
1030 | elif url_components.scheme.lower() == 'socks4a': |
1031 | socks_type = ProxyType.SOCKS4A | |
71aff188 | 1032 | |
cdd94c2e YCH |
1033 | def unquote_if_non_empty(s): |
1034 | if not s: | |
1035 | return s | |
1036 | return compat_urllib_parse_unquote_plus(s) | |
1037 | ||
71aff188 YCH |
1038 | proxy_args = ( |
1039 | socks_type, | |
1040 | url_components.hostname, url_components.port or 1080, | |
1041 | True, # Remote DNS | |
cdd94c2e YCH |
1042 | unquote_if_non_empty(url_components.username), |
1043 | unquote_if_non_empty(url_components.password), | |
71aff188 YCH |
1044 | ) |
1045 | ||
1046 | class SocksConnection(base_class): | |
1047 | def connect(self): | |
1048 | self.sock = sockssocket() | |
1049 | self.sock.setproxy(*proxy_args) | |
1050 | if type(self.timeout) in (int, float): | |
1051 | self.sock.settimeout(self.timeout) | |
1052 | self.sock.connect((self.host, self.port)) | |
1053 | ||
1054 | if isinstance(self, compat_http_client.HTTPSConnection): | |
1055 | if hasattr(self, '_context'): # Python > 2.6 | |
1056 | self.sock = self._context.wrap_socket( | |
1057 | self.sock, server_hostname=self.host) | |
1058 | else: | |
1059 | self.sock = ssl.wrap_socket(self.sock) | |
1060 | ||
1061 | return SocksConnection | |
1062 | ||
1063 | ||
be4a824d PH |
1064 | class YoutubeDLHTTPSHandler(compat_urllib_request.HTTPSHandler): |
1065 | def __init__(self, params, https_conn_class=None, *args, **kwargs): | |
1066 | compat_urllib_request.HTTPSHandler.__init__(self, *args, **kwargs) | |
1067 | self._https_conn_class = https_conn_class or compat_http_client.HTTPSConnection | |
1068 | self._params = params | |
1069 | ||
1070 | def https_open(self, req): | |
4f264c02 | 1071 | kwargs = {} |
71aff188 YCH |
1072 | conn_class = self._https_conn_class |
1073 | ||
4f264c02 JMF |
1074 | if hasattr(self, '_context'): # python > 2.6 |
1075 | kwargs['context'] = self._context | |
1076 | if hasattr(self, '_check_hostname'): # python 3.x | |
1077 | kwargs['check_hostname'] = self._check_hostname | |
71aff188 YCH |
1078 | |
1079 | socks_proxy = req.headers.get('Ytdl-socks-proxy') | |
1080 | if socks_proxy: | |
1081 | conn_class = make_socks_conn_class(conn_class, socks_proxy) | |
1082 | del req.headers['Ytdl-socks-proxy'] | |
1083 | ||
be4a824d | 1084 | return self.do_open(functools.partial( |
71aff188 | 1085 | _create_http_connection, self, conn_class, True), |
4f264c02 | 1086 | req, **kwargs) |
be4a824d PH |
1087 | |
1088 | ||
a6420bf5 S |
1089 | class YoutubeDLCookieProcessor(compat_urllib_request.HTTPCookieProcessor): |
1090 | def __init__(self, cookiejar=None): | |
1091 | compat_urllib_request.HTTPCookieProcessor.__init__(self, cookiejar) | |
1092 | ||
1093 | def http_response(self, request, response): | |
1094 | # Python 2 will choke on next HTTP request in row if there are non-ASCII | |
1095 | # characters in Set-Cookie HTTP header of last response (see | |
1096 | # https://github.com/rg3/youtube-dl/issues/6769). | |
1097 | # In order to at least prevent crashing we will percent encode Set-Cookie | |
1098 | # header before HTTPCookieProcessor starts processing it. | |
e28034c5 S |
1099 | # if sys.version_info < (3, 0) and response.headers: |
1100 | # for set_cookie_header in ('Set-Cookie', 'Set-Cookie2'): | |
1101 | # set_cookie = response.headers.get(set_cookie_header) | |
1102 | # if set_cookie: | |
1103 | # set_cookie_escaped = compat_urllib_parse.quote(set_cookie, b"%/;:@&=+$,!~*'()?#[] ") | |
1104 | # if set_cookie != set_cookie_escaped: | |
1105 | # del response.headers[set_cookie_header] | |
1106 | # response.headers[set_cookie_header] = set_cookie_escaped | |
a6420bf5 S |
1107 | return compat_urllib_request.HTTPCookieProcessor.http_response(self, request, response) |
1108 | ||
1109 | https_request = compat_urllib_request.HTTPCookieProcessor.http_request | |
1110 | https_response = http_response | |
1111 | ||
1112 | ||
46f59e89 S |
1113 | def extract_timezone(date_str): |
1114 | m = re.search( | |
1115 | r'^.{8,}?(?P<tz>Z$| ?(?P<sign>\+|-)(?P<hours>[0-9]{2}):?(?P<minutes>[0-9]{2})$)', | |
1116 | date_str) | |
1117 | if not m: | |
1118 | timezone = datetime.timedelta() | |
1119 | else: | |
1120 | date_str = date_str[:-len(m.group('tz'))] | |
1121 | if not m.group('sign'): | |
1122 | timezone = datetime.timedelta() | |
1123 | else: | |
1124 | sign = 1 if m.group('sign') == '+' else -1 | |
1125 | timezone = datetime.timedelta( | |
1126 | hours=sign * int(m.group('hours')), | |
1127 | minutes=sign * int(m.group('minutes'))) | |
1128 | return timezone, date_str | |
1129 | ||
1130 | ||
08b38d54 | 1131 | def parse_iso8601(date_str, delimiter='T', timezone=None): |
912b38b4 PH |
1132 | """ Return a UNIX timestamp from the given date """ |
1133 | ||
1134 | if date_str is None: | |
1135 | return None | |
1136 | ||
52c3a6e4 S |
1137 | date_str = re.sub(r'\.[0-9]+', '', date_str) |
1138 | ||
08b38d54 | 1139 | if timezone is None: |
46f59e89 S |
1140 | timezone, date_str = extract_timezone(date_str) |
1141 | ||
52c3a6e4 S |
1142 | try: |
1143 | date_format = '%Y-%m-%d{0}%H:%M:%S'.format(delimiter) | |
1144 | dt = datetime.datetime.strptime(date_str, date_format) - timezone | |
1145 | return calendar.timegm(dt.timetuple()) | |
1146 | except ValueError: | |
1147 | pass | |
912b38b4 PH |
1148 | |
1149 | ||
46f59e89 S |
1150 | def date_formats(day_first=True): |
1151 | return DATE_FORMATS_DAY_FIRST if day_first else DATE_FORMATS_MONTH_FIRST | |
1152 | ||
1153 | ||
42bdd9d0 | 1154 | def unified_strdate(date_str, day_first=True): |
bf50b038 | 1155 | """Return a string with the date in the format YYYYMMDD""" |
64e7ad60 PH |
1156 | |
1157 | if date_str is None: | |
1158 | return None | |
bf50b038 | 1159 | upload_date = None |
5f6a1245 | 1160 | # Replace commas |
026fcc04 | 1161 | date_str = date_str.replace(',', ' ') |
42bdd9d0 | 1162 | # Remove AM/PM + timezone |
9bb8e0a3 | 1163 | date_str = re.sub(r'(?i)\s*(?:AM|PM)(?:\s+[A-Z]+)?', '', date_str) |
46f59e89 | 1164 | _, date_str = extract_timezone(date_str) |
42bdd9d0 | 1165 | |
46f59e89 | 1166 | for expression in date_formats(day_first): |
bf50b038 JMF |
1167 | try: |
1168 | upload_date = datetime.datetime.strptime(date_str, expression).strftime('%Y%m%d') | |
5de90176 | 1169 | except ValueError: |
bf50b038 | 1170 | pass |
42393ce2 PH |
1171 | if upload_date is None: |
1172 | timetuple = email.utils.parsedate_tz(date_str) | |
1173 | if timetuple: | |
c6b9cf05 S |
1174 | try: |
1175 | upload_date = datetime.datetime(*timetuple[:6]).strftime('%Y%m%d') | |
1176 | except ValueError: | |
1177 | pass | |
6a750402 JMF |
1178 | if upload_date is not None: |
1179 | return compat_str(upload_date) | |
bf50b038 | 1180 | |
5f6a1245 | 1181 | |
46f59e89 S |
1182 | def unified_timestamp(date_str, day_first=True): |
1183 | if date_str is None: | |
1184 | return None | |
1185 | ||
2ae2ffda | 1186 | date_str = re.sub(r'[,|]', '', date_str) |
46f59e89 | 1187 | |
7dc2a74e | 1188 | pm_delta = 12 if re.search(r'(?i)PM', date_str) else 0 |
46f59e89 S |
1189 | timezone, date_str = extract_timezone(date_str) |
1190 | ||
1191 | # Remove AM/PM + timezone | |
1192 | date_str = re.sub(r'(?i)\s*(?:AM|PM)(?:\s+[A-Z]+)?', '', date_str) | |
1193 | ||
deef3195 S |
1194 | # Remove unrecognized timezones from ISO 8601 alike timestamps |
1195 | m = re.search(r'\d{1,2}:\d{1,2}(?:\.\d+)?(?P<tz>\s*[A-Z]+)$', date_str) | |
1196 | if m: | |
1197 | date_str = date_str[:-len(m.group('tz'))] | |
1198 | ||
46f59e89 S |
1199 | for expression in date_formats(day_first): |
1200 | try: | |
7dc2a74e | 1201 | dt = datetime.datetime.strptime(date_str, expression) - timezone + datetime.timedelta(hours=pm_delta) |
46f59e89 S |
1202 | return calendar.timegm(dt.timetuple()) |
1203 | except ValueError: | |
1204 | pass | |
1205 | timetuple = email.utils.parsedate_tz(date_str) | |
1206 | if timetuple: | |
7dc2a74e | 1207 | return calendar.timegm(timetuple) + pm_delta * 3600 |
46f59e89 S |
1208 | |
1209 | ||
28e614de | 1210 | def determine_ext(url, default_ext='unknown_video'): |
f4776371 S |
1211 | if url is None: |
1212 | return default_ext | |
9cb9a5df | 1213 | guess = url.partition('?')[0].rpartition('.')[2] |
73e79f2a PH |
1214 | if re.match(r'^[A-Za-z0-9]+$', guess): |
1215 | return guess | |
a7aaa398 S |
1216 | # Try extract ext from URLs like http://example.com/foo/bar.mp4/?download |
1217 | elif guess.rstrip('/') in KNOWN_EXTENSIONS: | |
9cb9a5df | 1218 | return guess.rstrip('/') |
73e79f2a | 1219 | else: |
cbdbb766 | 1220 | return default_ext |
73e79f2a | 1221 | |
5f6a1245 | 1222 | |
d4051a8e | 1223 | def subtitles_filename(filename, sub_lang, sub_format): |
28e614de | 1224 | return filename.rsplit('.', 1)[0] + '.' + sub_lang + '.' + sub_format |
d4051a8e | 1225 | |
5f6a1245 | 1226 | |
bd558525 | 1227 | def date_from_str(date_str): |
37254abc JMF |
1228 | """ |
1229 | Return a datetime object from a string in the format YYYYMMDD or | |
1230 | (now|today)[+-][0-9](day|week|month|year)(s)?""" | |
1231 | today = datetime.date.today() | |
f8795e10 | 1232 | if date_str in ('now', 'today'): |
37254abc | 1233 | return today |
f8795e10 PH |
1234 | if date_str == 'yesterday': |
1235 | return today - datetime.timedelta(days=1) | |
ec85ded8 | 1236 | match = re.match(r'(now|today)(?P<sign>[+-])(?P<time>\d+)(?P<unit>day|week|month|year)(s)?', date_str) |
37254abc JMF |
1237 | if match is not None: |
1238 | sign = match.group('sign') | |
1239 | time = int(match.group('time')) | |
1240 | if sign == '-': | |
1241 | time = -time | |
1242 | unit = match.group('unit') | |
dfb1b146 | 1243 | # A bad approximation? |
37254abc JMF |
1244 | if unit == 'month': |
1245 | unit = 'day' | |
1246 | time *= 30 | |
1247 | elif unit == 'year': | |
1248 | unit = 'day' | |
1249 | time *= 365 | |
1250 | unit += 's' | |
1251 | delta = datetime.timedelta(**{unit: time}) | |
1252 | return today + delta | |
611c1dd9 | 1253 | return datetime.datetime.strptime(date_str, '%Y%m%d').date() |
5f6a1245 JW |
1254 | |
1255 | ||
e63fc1be | 1256 | def hyphenate_date(date_str): |
1257 | """ | |
1258 | Convert a date in 'YYYYMMDD' format to 'YYYY-MM-DD' format""" | |
1259 | match = re.match(r'^(\d\d\d\d)(\d\d)(\d\d)$', date_str) | |
1260 | if match is not None: | |
1261 | return '-'.join(match.groups()) | |
1262 | else: | |
1263 | return date_str | |
1264 | ||
5f6a1245 | 1265 | |
bd558525 JMF |
1266 | class DateRange(object): |
1267 | """Represents a time interval between two dates""" | |
5f6a1245 | 1268 | |
bd558525 JMF |
1269 | def __init__(self, start=None, end=None): |
1270 | """start and end must be strings in the format accepted by date""" | |
1271 | if start is not None: | |
1272 | self.start = date_from_str(start) | |
1273 | else: | |
1274 | self.start = datetime.datetime.min.date() | |
1275 | if end is not None: | |
1276 | self.end = date_from_str(end) | |
1277 | else: | |
1278 | self.end = datetime.datetime.max.date() | |
37254abc | 1279 | if self.start > self.end: |
bd558525 | 1280 | raise ValueError('Date range: "%s" , the start date must be before the end date' % self) |
5f6a1245 | 1281 | |
bd558525 JMF |
1282 | @classmethod |
1283 | def day(cls, day): | |
1284 | """Returns a range that only contains the given day""" | |
5f6a1245 JW |
1285 | return cls(day, day) |
1286 | ||
bd558525 JMF |
1287 | def __contains__(self, date): |
1288 | """Check if the date is in the range""" | |
37254abc JMF |
1289 | if not isinstance(date, datetime.date): |
1290 | date = date_from_str(date) | |
1291 | return self.start <= date <= self.end | |
5f6a1245 | 1292 | |
bd558525 | 1293 | def __str__(self): |
5f6a1245 | 1294 | return '%s - %s' % (self.start.isoformat(), self.end.isoformat()) |
c496ca96 PH |
1295 | |
1296 | ||
1297 | def platform_name(): | |
1298 | """ Returns the platform name as a compat_str """ | |
1299 | res = platform.platform() | |
1300 | if isinstance(res, bytes): | |
1301 | res = res.decode(preferredencoding()) | |
1302 | ||
1303 | assert isinstance(res, compat_str) | |
1304 | return res | |
c257baff PH |
1305 | |
1306 | ||
b58ddb32 PH |
1307 | def _windows_write_string(s, out): |
1308 | """ Returns True if the string was written using special methods, | |
1309 | False if it has yet to be written out.""" | |
1310 | # Adapted from http://stackoverflow.com/a/3259271/35070 | |
1311 | ||
1312 | import ctypes | |
1313 | import ctypes.wintypes | |
1314 | ||
1315 | WIN_OUTPUT_IDS = { | |
1316 | 1: -11, | |
1317 | 2: -12, | |
1318 | } | |
1319 | ||
a383a98a PH |
1320 | try: |
1321 | fileno = out.fileno() | |
1322 | except AttributeError: | |
1323 | # If the output stream doesn't have a fileno, it's virtual | |
1324 | return False | |
aa42e873 PH |
1325 | except io.UnsupportedOperation: |
1326 | # Some strange Windows pseudo files? | |
1327 | return False | |
b58ddb32 PH |
1328 | if fileno not in WIN_OUTPUT_IDS: |
1329 | return False | |
1330 | ||
e2f89ec7 | 1331 | GetStdHandle = ctypes.WINFUNCTYPE( |
b58ddb32 | 1332 | ctypes.wintypes.HANDLE, ctypes.wintypes.DWORD)( |
611c1dd9 | 1333 | (b'GetStdHandle', ctypes.windll.kernel32)) |
b58ddb32 PH |
1334 | h = GetStdHandle(WIN_OUTPUT_IDS[fileno]) |
1335 | ||
e2f89ec7 | 1336 | WriteConsoleW = ctypes.WINFUNCTYPE( |
b58ddb32 PH |
1337 | ctypes.wintypes.BOOL, ctypes.wintypes.HANDLE, ctypes.wintypes.LPWSTR, |
1338 | ctypes.wintypes.DWORD, ctypes.POINTER(ctypes.wintypes.DWORD), | |
611c1dd9 | 1339 | ctypes.wintypes.LPVOID)((b'WriteConsoleW', ctypes.windll.kernel32)) |
b58ddb32 PH |
1340 | written = ctypes.wintypes.DWORD(0) |
1341 | ||
611c1dd9 | 1342 | GetFileType = ctypes.WINFUNCTYPE(ctypes.wintypes.DWORD, ctypes.wintypes.DWORD)((b'GetFileType', ctypes.windll.kernel32)) |
b58ddb32 PH |
1343 | FILE_TYPE_CHAR = 0x0002 |
1344 | FILE_TYPE_REMOTE = 0x8000 | |
e2f89ec7 | 1345 | GetConsoleMode = ctypes.WINFUNCTYPE( |
b58ddb32 PH |
1346 | ctypes.wintypes.BOOL, ctypes.wintypes.HANDLE, |
1347 | ctypes.POINTER(ctypes.wintypes.DWORD))( | |
611c1dd9 | 1348 | (b'GetConsoleMode', ctypes.windll.kernel32)) |
b58ddb32 PH |
1349 | INVALID_HANDLE_VALUE = ctypes.wintypes.DWORD(-1).value |
1350 | ||
1351 | def not_a_console(handle): | |
1352 | if handle == INVALID_HANDLE_VALUE or handle is None: | |
1353 | return True | |
8fb3ac36 PH |
1354 | return ((GetFileType(handle) & ~FILE_TYPE_REMOTE) != FILE_TYPE_CHAR or |
1355 | GetConsoleMode(handle, ctypes.byref(ctypes.wintypes.DWORD())) == 0) | |
b58ddb32 PH |
1356 | |
1357 | if not_a_console(h): | |
1358 | return False | |
1359 | ||
d1b9c912 PH |
1360 | def next_nonbmp_pos(s): |
1361 | try: | |
1362 | return next(i for i, c in enumerate(s) if ord(c) > 0xffff) | |
1363 | except StopIteration: | |
1364 | return len(s) | |
1365 | ||
1366 | while s: | |
1367 | count = min(next_nonbmp_pos(s), 1024) | |
1368 | ||
b58ddb32 | 1369 | ret = WriteConsoleW( |
d1b9c912 | 1370 | h, s, count if count else 2, ctypes.byref(written), None) |
b58ddb32 PH |
1371 | if ret == 0: |
1372 | raise OSError('Failed to write string') | |
d1b9c912 PH |
1373 | if not count: # We just wrote a non-BMP character |
1374 | assert written.value == 2 | |
1375 | s = s[1:] | |
1376 | else: | |
1377 | assert written.value > 0 | |
1378 | s = s[written.value:] | |
b58ddb32 PH |
1379 | return True |
1380 | ||
1381 | ||
734f90bb | 1382 | def write_string(s, out=None, encoding=None): |
7459e3a2 PH |
1383 | if out is None: |
1384 | out = sys.stderr | |
8bf48f23 | 1385 | assert type(s) == compat_str |
7459e3a2 | 1386 | |
b58ddb32 PH |
1387 | if sys.platform == 'win32' and encoding is None and hasattr(out, 'fileno'): |
1388 | if _windows_write_string(s, out): | |
1389 | return | |
1390 | ||
7459e3a2 PH |
1391 | if ('b' in getattr(out, 'mode', '') or |
1392 | sys.version_info[0] < 3): # Python 2 lies about mode of sys.stderr | |
104aa738 PH |
1393 | byt = s.encode(encoding or preferredencoding(), 'ignore') |
1394 | out.write(byt) | |
1395 | elif hasattr(out, 'buffer'): | |
1396 | enc = encoding or getattr(out, 'encoding', None) or preferredencoding() | |
1397 | byt = s.encode(enc, 'ignore') | |
1398 | out.buffer.write(byt) | |
1399 | else: | |
8bf48f23 | 1400 | out.write(s) |
7459e3a2 PH |
1401 | out.flush() |
1402 | ||
1403 | ||
48ea9cea PH |
1404 | def bytes_to_intlist(bs): |
1405 | if not bs: | |
1406 | return [] | |
1407 | if isinstance(bs[0], int): # Python 3 | |
1408 | return list(bs) | |
1409 | else: | |
1410 | return [ord(c) for c in bs] | |
1411 | ||
c257baff | 1412 | |
cba892fa | 1413 | def intlist_to_bytes(xs): |
1414 | if not xs: | |
1415 | return b'' | |
edaa23f8 | 1416 | return compat_struct_pack('%dB' % len(xs), *xs) |
c38b1e77 PH |
1417 | |
1418 | ||
c1c9a79c PH |
1419 | # Cross-platform file locking |
1420 | if sys.platform == 'win32': | |
1421 | import ctypes.wintypes | |
1422 | import msvcrt | |
1423 | ||
1424 | class OVERLAPPED(ctypes.Structure): | |
1425 | _fields_ = [ | |
1426 | ('Internal', ctypes.wintypes.LPVOID), | |
1427 | ('InternalHigh', ctypes.wintypes.LPVOID), | |
1428 | ('Offset', ctypes.wintypes.DWORD), | |
1429 | ('OffsetHigh', ctypes.wintypes.DWORD), | |
1430 | ('hEvent', ctypes.wintypes.HANDLE), | |
1431 | ] | |
1432 | ||
1433 | kernel32 = ctypes.windll.kernel32 | |
1434 | LockFileEx = kernel32.LockFileEx | |
1435 | LockFileEx.argtypes = [ | |
1436 | ctypes.wintypes.HANDLE, # hFile | |
1437 | ctypes.wintypes.DWORD, # dwFlags | |
1438 | ctypes.wintypes.DWORD, # dwReserved | |
1439 | ctypes.wintypes.DWORD, # nNumberOfBytesToLockLow | |
1440 | ctypes.wintypes.DWORD, # nNumberOfBytesToLockHigh | |
1441 | ctypes.POINTER(OVERLAPPED) # Overlapped | |
1442 | ] | |
1443 | LockFileEx.restype = ctypes.wintypes.BOOL | |
1444 | UnlockFileEx = kernel32.UnlockFileEx | |
1445 | UnlockFileEx.argtypes = [ | |
1446 | ctypes.wintypes.HANDLE, # hFile | |
1447 | ctypes.wintypes.DWORD, # dwReserved | |
1448 | ctypes.wintypes.DWORD, # nNumberOfBytesToLockLow | |
1449 | ctypes.wintypes.DWORD, # nNumberOfBytesToLockHigh | |
1450 | ctypes.POINTER(OVERLAPPED) # Overlapped | |
1451 | ] | |
1452 | UnlockFileEx.restype = ctypes.wintypes.BOOL | |
1453 | whole_low = 0xffffffff | |
1454 | whole_high = 0x7fffffff | |
1455 | ||
1456 | def _lock_file(f, exclusive): | |
1457 | overlapped = OVERLAPPED() | |
1458 | overlapped.Offset = 0 | |
1459 | overlapped.OffsetHigh = 0 | |
1460 | overlapped.hEvent = 0 | |
1461 | f._lock_file_overlapped_p = ctypes.pointer(overlapped) | |
1462 | handle = msvcrt.get_osfhandle(f.fileno()) | |
1463 | if not LockFileEx(handle, 0x2 if exclusive else 0x0, 0, | |
1464 | whole_low, whole_high, f._lock_file_overlapped_p): | |
1465 | raise OSError('Locking file failed: %r' % ctypes.FormatError()) | |
1466 | ||
1467 | def _unlock_file(f): | |
1468 | assert f._lock_file_overlapped_p | |
1469 | handle = msvcrt.get_osfhandle(f.fileno()) | |
1470 | if not UnlockFileEx(handle, 0, | |
1471 | whole_low, whole_high, f._lock_file_overlapped_p): | |
1472 | raise OSError('Unlocking file failed: %r' % ctypes.FormatError()) | |
1473 | ||
1474 | else: | |
399a76e6 YCH |
1475 | # Some platforms, such as Jython, is missing fcntl |
1476 | try: | |
1477 | import fcntl | |
c1c9a79c | 1478 | |
399a76e6 YCH |
1479 | def _lock_file(f, exclusive): |
1480 | fcntl.flock(f, fcntl.LOCK_EX if exclusive else fcntl.LOCK_SH) | |
c1c9a79c | 1481 | |
399a76e6 YCH |
1482 | def _unlock_file(f): |
1483 | fcntl.flock(f, fcntl.LOCK_UN) | |
1484 | except ImportError: | |
1485 | UNSUPPORTED_MSG = 'file locking is not supported on this platform' | |
1486 | ||
1487 | def _lock_file(f, exclusive): | |
1488 | raise IOError(UNSUPPORTED_MSG) | |
1489 | ||
1490 | def _unlock_file(f): | |
1491 | raise IOError(UNSUPPORTED_MSG) | |
c1c9a79c PH |
1492 | |
1493 | ||
1494 | class locked_file(object): | |
1495 | def __init__(self, filename, mode, encoding=None): | |
1496 | assert mode in ['r', 'a', 'w'] | |
1497 | self.f = io.open(filename, mode, encoding=encoding) | |
1498 | self.mode = mode | |
1499 | ||
1500 | def __enter__(self): | |
1501 | exclusive = self.mode != 'r' | |
1502 | try: | |
1503 | _lock_file(self.f, exclusive) | |
1504 | except IOError: | |
1505 | self.f.close() | |
1506 | raise | |
1507 | return self | |
1508 | ||
1509 | def __exit__(self, etype, value, traceback): | |
1510 | try: | |
1511 | _unlock_file(self.f) | |
1512 | finally: | |
1513 | self.f.close() | |
1514 | ||
1515 | def __iter__(self): | |
1516 | return iter(self.f) | |
1517 | ||
1518 | def write(self, *args): | |
1519 | return self.f.write(*args) | |
1520 | ||
1521 | def read(self, *args): | |
1522 | return self.f.read(*args) | |
4eb7f1d1 JMF |
1523 | |
1524 | ||
4644ac55 S |
1525 | def get_filesystem_encoding(): |
1526 | encoding = sys.getfilesystemencoding() | |
1527 | return encoding if encoding is not None else 'utf-8' | |
1528 | ||
1529 | ||
4eb7f1d1 | 1530 | def shell_quote(args): |
a6a173c2 | 1531 | quoted_args = [] |
4644ac55 | 1532 | encoding = get_filesystem_encoding() |
a6a173c2 JMF |
1533 | for a in args: |
1534 | if isinstance(a, bytes): | |
1535 | # We may get a filename encoded with 'encodeFilename' | |
1536 | a = a.decode(encoding) | |
aefce8e6 | 1537 | quoted_args.append(compat_shlex_quote(a)) |
28e614de | 1538 | return ' '.join(quoted_args) |
9d4660ca PH |
1539 | |
1540 | ||
1541 | def smuggle_url(url, data): | |
1542 | """ Pass additional data in a URL for internal use. """ | |
1543 | ||
81953d1a RA |
1544 | url, idata = unsmuggle_url(url, {}) |
1545 | data.update(idata) | |
15707c7e | 1546 | sdata = compat_urllib_parse_urlencode( |
28e614de PH |
1547 | {'__youtubedl_smuggle': json.dumps(data)}) |
1548 | return url + '#' + sdata | |
9d4660ca PH |
1549 | |
1550 | ||
79f82953 | 1551 | def unsmuggle_url(smug_url, default=None): |
83e865a3 | 1552 | if '#__youtubedl_smuggle' not in smug_url: |
79f82953 | 1553 | return smug_url, default |
28e614de PH |
1554 | url, _, sdata = smug_url.rpartition('#') |
1555 | jsond = compat_parse_qs(sdata)['__youtubedl_smuggle'][0] | |
9d4660ca PH |
1556 | data = json.loads(jsond) |
1557 | return url, data | |
02dbf93f PH |
1558 | |
1559 | ||
02dbf93f PH |
1560 | def format_bytes(bytes): |
1561 | if bytes is None: | |
28e614de | 1562 | return 'N/A' |
02dbf93f PH |
1563 | if type(bytes) is str: |
1564 | bytes = float(bytes) | |
1565 | if bytes == 0.0: | |
1566 | exponent = 0 | |
1567 | else: | |
1568 | exponent = int(math.log(bytes, 1024.0)) | |
28e614de | 1569 | suffix = ['B', 'KiB', 'MiB', 'GiB', 'TiB', 'PiB', 'EiB', 'ZiB', 'YiB'][exponent] |
02dbf93f | 1570 | converted = float(bytes) / float(1024 ** exponent) |
28e614de | 1571 | return '%.2f%s' % (converted, suffix) |
f53c966a | 1572 | |
1c088fa8 | 1573 | |
fb47597b S |
1574 | def lookup_unit_table(unit_table, s): |
1575 | units_re = '|'.join(re.escape(u) for u in unit_table) | |
1576 | m = re.match( | |
782b1b5b | 1577 | r'(?P<num>[0-9]+(?:[,.][0-9]*)?)\s*(?P<unit>%s)\b' % units_re, s) |
fb47597b S |
1578 | if not m: |
1579 | return None | |
1580 | num_str = m.group('num').replace(',', '.') | |
1581 | mult = unit_table[m.group('unit')] | |
1582 | return int(float(num_str) * mult) | |
1583 | ||
1584 | ||
be64b5b0 PH |
1585 | def parse_filesize(s): |
1586 | if s is None: | |
1587 | return None | |
1588 | ||
dfb1b146 | 1589 | # The lower-case forms are of course incorrect and unofficial, |
be64b5b0 PH |
1590 | # but we support those too |
1591 | _UNIT_TABLE = { | |
1592 | 'B': 1, | |
1593 | 'b': 1, | |
70852b47 | 1594 | 'bytes': 1, |
be64b5b0 PH |
1595 | 'KiB': 1024, |
1596 | 'KB': 1000, | |
1597 | 'kB': 1024, | |
1598 | 'Kb': 1000, | |
13585d76 | 1599 | 'kb': 1000, |
70852b47 YCH |
1600 | 'kilobytes': 1000, |
1601 | 'kibibytes': 1024, | |
be64b5b0 PH |
1602 | 'MiB': 1024 ** 2, |
1603 | 'MB': 1000 ** 2, | |
1604 | 'mB': 1024 ** 2, | |
1605 | 'Mb': 1000 ** 2, | |
13585d76 | 1606 | 'mb': 1000 ** 2, |
70852b47 YCH |
1607 | 'megabytes': 1000 ** 2, |
1608 | 'mebibytes': 1024 ** 2, | |
be64b5b0 PH |
1609 | 'GiB': 1024 ** 3, |
1610 | 'GB': 1000 ** 3, | |
1611 | 'gB': 1024 ** 3, | |
1612 | 'Gb': 1000 ** 3, | |
13585d76 | 1613 | 'gb': 1000 ** 3, |
70852b47 YCH |
1614 | 'gigabytes': 1000 ** 3, |
1615 | 'gibibytes': 1024 ** 3, | |
be64b5b0 PH |
1616 | 'TiB': 1024 ** 4, |
1617 | 'TB': 1000 ** 4, | |
1618 | 'tB': 1024 ** 4, | |
1619 | 'Tb': 1000 ** 4, | |
13585d76 | 1620 | 'tb': 1000 ** 4, |
70852b47 YCH |
1621 | 'terabytes': 1000 ** 4, |
1622 | 'tebibytes': 1024 ** 4, | |
be64b5b0 PH |
1623 | 'PiB': 1024 ** 5, |
1624 | 'PB': 1000 ** 5, | |
1625 | 'pB': 1024 ** 5, | |
1626 | 'Pb': 1000 ** 5, | |
13585d76 | 1627 | 'pb': 1000 ** 5, |
70852b47 YCH |
1628 | 'petabytes': 1000 ** 5, |
1629 | 'pebibytes': 1024 ** 5, | |
be64b5b0 PH |
1630 | 'EiB': 1024 ** 6, |
1631 | 'EB': 1000 ** 6, | |
1632 | 'eB': 1024 ** 6, | |
1633 | 'Eb': 1000 ** 6, | |
13585d76 | 1634 | 'eb': 1000 ** 6, |
70852b47 YCH |
1635 | 'exabytes': 1000 ** 6, |
1636 | 'exbibytes': 1024 ** 6, | |
be64b5b0 PH |
1637 | 'ZiB': 1024 ** 7, |
1638 | 'ZB': 1000 ** 7, | |
1639 | 'zB': 1024 ** 7, | |
1640 | 'Zb': 1000 ** 7, | |
13585d76 | 1641 | 'zb': 1000 ** 7, |
70852b47 YCH |
1642 | 'zettabytes': 1000 ** 7, |
1643 | 'zebibytes': 1024 ** 7, | |
be64b5b0 PH |
1644 | 'YiB': 1024 ** 8, |
1645 | 'YB': 1000 ** 8, | |
1646 | 'yB': 1024 ** 8, | |
1647 | 'Yb': 1000 ** 8, | |
13585d76 | 1648 | 'yb': 1000 ** 8, |
70852b47 YCH |
1649 | 'yottabytes': 1000 ** 8, |
1650 | 'yobibytes': 1024 ** 8, | |
be64b5b0 PH |
1651 | } |
1652 | ||
fb47597b S |
1653 | return lookup_unit_table(_UNIT_TABLE, s) |
1654 | ||
1655 | ||
1656 | def parse_count(s): | |
1657 | if s is None: | |
be64b5b0 PH |
1658 | return None |
1659 | ||
fb47597b S |
1660 | s = s.strip() |
1661 | ||
1662 | if re.match(r'^[\d,.]+$', s): | |
1663 | return str_to_int(s) | |
1664 | ||
1665 | _UNIT_TABLE = { | |
1666 | 'k': 1000, | |
1667 | 'K': 1000, | |
1668 | 'm': 1000 ** 2, | |
1669 | 'M': 1000 ** 2, | |
1670 | 'kk': 1000 ** 2, | |
1671 | 'KK': 1000 ** 2, | |
1672 | } | |
be64b5b0 | 1673 | |
fb47597b | 1674 | return lookup_unit_table(_UNIT_TABLE, s) |
be64b5b0 | 1675 | |
2f7ae819 | 1676 | |
a942d6cb | 1677 | def month_by_name(name, lang='en'): |
caefb1de PH |
1678 | """ Return the number of a month by (locale-independently) English name """ |
1679 | ||
f6717dec | 1680 | month_names = MONTH_NAMES.get(lang, MONTH_NAMES['en']) |
a942d6cb | 1681 | |
caefb1de | 1682 | try: |
f6717dec | 1683 | return month_names.index(name) + 1 |
7105440c YCH |
1684 | except ValueError: |
1685 | return None | |
1686 | ||
1687 | ||
1688 | def month_by_abbreviation(abbrev): | |
1689 | """ Return the number of a month by (locale-independently) English | |
1690 | abbreviations """ | |
1691 | ||
1692 | try: | |
1693 | return [s[:3] for s in ENGLISH_MONTH_NAMES].index(abbrev) + 1 | |
caefb1de PH |
1694 | except ValueError: |
1695 | return None | |
18258362 JMF |
1696 | |
1697 | ||
5aafe895 | 1698 | def fix_xml_ampersands(xml_str): |
18258362 | 1699 | """Replace all the '&' by '&' in XML""" |
5aafe895 PH |
1700 | return re.sub( |
1701 | r'&(?!amp;|lt;|gt;|apos;|quot;|#x[0-9a-fA-F]{,4};|#[0-9]{,4};)', | |
28e614de | 1702 | '&', |
5aafe895 | 1703 | xml_str) |
e3946f98 PH |
1704 | |
1705 | ||
1706 | def setproctitle(title): | |
8bf48f23 | 1707 | assert isinstance(title, compat_str) |
c1c05c67 YCH |
1708 | |
1709 | # ctypes in Jython is not complete | |
1710 | # http://bugs.jython.org/issue2148 | |
1711 | if sys.platform.startswith('java'): | |
1712 | return | |
1713 | ||
e3946f98 | 1714 | try: |
611c1dd9 | 1715 | libc = ctypes.cdll.LoadLibrary('libc.so.6') |
e3946f98 PH |
1716 | except OSError: |
1717 | return | |
2f49bcd6 RC |
1718 | except TypeError: |
1719 | # LoadLibrary in Windows Python 2.7.13 only expects | |
1720 | # a bytestring, but since unicode_literals turns | |
1721 | # every string into a unicode string, it fails. | |
1722 | return | |
6eefe533 PH |
1723 | title_bytes = title.encode('utf-8') |
1724 | buf = ctypes.create_string_buffer(len(title_bytes)) | |
1725 | buf.value = title_bytes | |
e3946f98 | 1726 | try: |
6eefe533 | 1727 | libc.prctl(15, buf, 0, 0, 0) |
e3946f98 PH |
1728 | except AttributeError: |
1729 | return # Strange libc, just skip this | |
d7dda168 PH |
1730 | |
1731 | ||
1732 | def remove_start(s, start): | |
46bc9b7d | 1733 | return s[len(start):] if s is not None and s.startswith(start) else s |
29eb5174 PH |
1734 | |
1735 | ||
2b9faf55 | 1736 | def remove_end(s, end): |
46bc9b7d | 1737 | return s[:-len(end)] if s is not None and s.endswith(end) else s |
2b9faf55 PH |
1738 | |
1739 | ||
31b2051e S |
1740 | def remove_quotes(s): |
1741 | if s is None or len(s) < 2: | |
1742 | return s | |
1743 | for quote in ('"', "'", ): | |
1744 | if s[0] == quote and s[-1] == quote: | |
1745 | return s[1:-1] | |
1746 | return s | |
1747 | ||
1748 | ||
29eb5174 | 1749 | def url_basename(url): |
9b8aaeed | 1750 | path = compat_urlparse.urlparse(url).path |
28e614de | 1751 | return path.strip('/').split('/')[-1] |
aa94a6d3 PH |
1752 | |
1753 | ||
02dc0a36 S |
1754 | def base_url(url): |
1755 | return re.match(r'https?://[^?#&]+/', url).group() | |
1756 | ||
1757 | ||
e34c3361 | 1758 | def urljoin(base, path): |
4b5de77b S |
1759 | if isinstance(path, bytes): |
1760 | path = path.decode('utf-8') | |
e34c3361 S |
1761 | if not isinstance(path, compat_str) or not path: |
1762 | return None | |
b0c65c67 | 1763 | if re.match(r'^(?:https?:)?//', path): |
e34c3361 | 1764 | return path |
4b5de77b S |
1765 | if isinstance(base, bytes): |
1766 | base = base.decode('utf-8') | |
1767 | if not isinstance(base, compat_str) or not re.match( | |
1768 | r'^(?:https?:)?//', base): | |
e34c3361 S |
1769 | return None |
1770 | return compat_urlparse.urljoin(base, path) | |
1771 | ||
1772 | ||
aa94a6d3 PH |
1773 | class HEADRequest(compat_urllib_request.Request): |
1774 | def get_method(self): | |
611c1dd9 | 1775 | return 'HEAD' |
7217e148 PH |
1776 | |
1777 | ||
95cf60e8 S |
1778 | class PUTRequest(compat_urllib_request.Request): |
1779 | def get_method(self): | |
1780 | return 'PUT' | |
1781 | ||
1782 | ||
9732d77e | 1783 | def int_or_none(v, scale=1, default=None, get_attr=None, invscale=1): |
28746fbd PH |
1784 | if get_attr: |
1785 | if v is not None: | |
1786 | v = getattr(v, get_attr, None) | |
9572013d PH |
1787 | if v == '': |
1788 | v = None | |
1812afb7 S |
1789 | if v is None: |
1790 | return default | |
1791 | try: | |
1792 | return int(v) * invscale // scale | |
1793 | except ValueError: | |
af98f8ff | 1794 | return default |
9732d77e | 1795 | |
9572013d | 1796 | |
40a90862 JMF |
1797 | def str_or_none(v, default=None): |
1798 | return default if v is None else compat_str(v) | |
1799 | ||
9732d77e PH |
1800 | |
1801 | def str_to_int(int_str): | |
48d4681e | 1802 | """ A more relaxed version of int_or_none """ |
9732d77e PH |
1803 | if int_str is None: |
1804 | return None | |
28e614de | 1805 | int_str = re.sub(r'[,\.\+]', '', int_str) |
9732d77e | 1806 | return int(int_str) |
608d11f5 PH |
1807 | |
1808 | ||
9732d77e | 1809 | def float_or_none(v, scale=1, invscale=1, default=None): |
caf80631 S |
1810 | if v is None: |
1811 | return default | |
1812 | try: | |
1813 | return float(v) * invscale / scale | |
1814 | except ValueError: | |
1815 | return default | |
43f775e4 PH |
1816 | |
1817 | ||
c7e327c4 S |
1818 | def bool_or_none(v, default=None): |
1819 | return v if isinstance(v, bool) else default | |
1820 | ||
1821 | ||
b72b4431 S |
1822 | def strip_or_none(v): |
1823 | return None if v is None else v.strip() | |
1824 | ||
1825 | ||
608d11f5 | 1826 | def parse_duration(s): |
8f9312c3 | 1827 | if not isinstance(s, compat_basestring): |
608d11f5 PH |
1828 | return None |
1829 | ||
ca7b3246 S |
1830 | s = s.strip() |
1831 | ||
acaff495 | 1832 | days, hours, mins, secs, ms = [None] * 5 |
15846398 | 1833 | m = re.match(r'(?:(?:(?:(?P<days>[0-9]+):)?(?P<hours>[0-9]+):)?(?P<mins>[0-9]+):)?(?P<secs>[0-9]+)(?P<ms>\.[0-9]+)?Z?$', s) |
acaff495 | 1834 | if m: |
1835 | days, hours, mins, secs, ms = m.groups() | |
1836 | else: | |
1837 | m = re.match( | |
1838 | r'''(?ix)(?:P?T)? | |
8f4b58d7 | 1839 | (?: |
acaff495 | 1840 | (?P<days>[0-9]+)\s*d(?:ays?)?\s* |
8f4b58d7 | 1841 | )? |
acaff495 | 1842 | (?: |
1843 | (?P<hours>[0-9]+)\s*h(?:ours?)?\s* | |
1844 | )? | |
1845 | (?: | |
1846 | (?P<mins>[0-9]+)\s*m(?:in(?:ute)?s?)?\s* | |
1847 | )? | |
1848 | (?: | |
1849 | (?P<secs>[0-9]+)(?P<ms>\.[0-9]+)?\s*s(?:ec(?:ond)?s?)?\s* | |
15846398 | 1850 | )?Z?$''', s) |
acaff495 | 1851 | if m: |
1852 | days, hours, mins, secs, ms = m.groups() | |
1853 | else: | |
15846398 | 1854 | m = re.match(r'(?i)(?:(?P<hours>[0-9.]+)\s*(?:hours?)|(?P<mins>[0-9.]+)\s*(?:mins?\.?|minutes?)\s*)Z?$', s) |
acaff495 | 1855 | if m: |
1856 | hours, mins = m.groups() | |
1857 | else: | |
1858 | return None | |
1859 | ||
1860 | duration = 0 | |
1861 | if secs: | |
1862 | duration += float(secs) | |
1863 | if mins: | |
1864 | duration += float(mins) * 60 | |
1865 | if hours: | |
1866 | duration += float(hours) * 60 * 60 | |
1867 | if days: | |
1868 | duration += float(days) * 24 * 60 * 60 | |
1869 | if ms: | |
1870 | duration += float(ms) | |
1871 | return duration | |
91d7d0b3 JMF |
1872 | |
1873 | ||
e65e4c88 | 1874 | def prepend_extension(filename, ext, expected_real_ext=None): |
5f6a1245 | 1875 | name, real_ext = os.path.splitext(filename) |
e65e4c88 S |
1876 | return ( |
1877 | '{0}.{1}{2}'.format(name, ext, real_ext) | |
1878 | if not expected_real_ext or real_ext[1:] == expected_real_ext | |
1879 | else '{0}.{1}'.format(filename, ext)) | |
d70ad093 PH |
1880 | |
1881 | ||
b3ed15b7 S |
1882 | def replace_extension(filename, ext, expected_real_ext=None): |
1883 | name, real_ext = os.path.splitext(filename) | |
1884 | return '{0}.{1}'.format( | |
1885 | name if not expected_real_ext or real_ext[1:] == expected_real_ext else filename, | |
1886 | ext) | |
1887 | ||
1888 | ||
d70ad093 PH |
1889 | def check_executable(exe, args=[]): |
1890 | """ Checks if the given binary is installed somewhere in PATH, and returns its name. | |
1891 | args can be a list of arguments for a short output (like -version) """ | |
1892 | try: | |
1893 | subprocess.Popen([exe] + args, stdout=subprocess.PIPE, stderr=subprocess.PIPE).communicate() | |
1894 | except OSError: | |
1895 | return False | |
1896 | return exe | |
b7ab0590 PH |
1897 | |
1898 | ||
95807118 | 1899 | def get_exe_version(exe, args=['--version'], |
cae97f65 | 1900 | version_re=None, unrecognized='present'): |
95807118 PH |
1901 | """ Returns the version of the specified executable, |
1902 | or False if the executable is not present """ | |
1903 | try: | |
b64d04c1 YCH |
1904 | # STDIN should be redirected too. On UNIX-like systems, ffmpeg triggers |
1905 | # SIGTTOU if youtube-dl is run in the background. | |
1906 | # See https://github.com/rg3/youtube-dl/issues/955#issuecomment-209789656 | |
cae97f65 | 1907 | out, _ = subprocess.Popen( |
54116803 | 1908 | [encodeArgument(exe)] + args, |
00ca7552 | 1909 | stdin=subprocess.PIPE, |
95807118 PH |
1910 | stdout=subprocess.PIPE, stderr=subprocess.STDOUT).communicate() |
1911 | except OSError: | |
1912 | return False | |
cae97f65 PH |
1913 | if isinstance(out, bytes): # Python 2.x |
1914 | out = out.decode('ascii', 'ignore') | |
1915 | return detect_exe_version(out, version_re, unrecognized) | |
1916 | ||
1917 | ||
1918 | def detect_exe_version(output, version_re=None, unrecognized='present'): | |
1919 | assert isinstance(output, compat_str) | |
1920 | if version_re is None: | |
1921 | version_re = r'version\s+([-0-9._a-zA-Z]+)' | |
1922 | m = re.search(version_re, output) | |
95807118 PH |
1923 | if m: |
1924 | return m.group(1) | |
1925 | else: | |
1926 | return unrecognized | |
1927 | ||
1928 | ||
b7ab0590 | 1929 | class PagedList(object): |
dd26ced1 PH |
1930 | def __len__(self): |
1931 | # This is only useful for tests | |
1932 | return len(self.getslice()) | |
1933 | ||
9c44d242 PH |
1934 | |
1935 | class OnDemandPagedList(PagedList): | |
b95dc034 | 1936 | def __init__(self, pagefunc, pagesize, use_cache=False): |
9c44d242 PH |
1937 | self._pagefunc = pagefunc |
1938 | self._pagesize = pagesize | |
b95dc034 YCH |
1939 | self._use_cache = use_cache |
1940 | if use_cache: | |
1941 | self._cache = {} | |
9c44d242 | 1942 | |
b7ab0590 PH |
1943 | def getslice(self, start=0, end=None): |
1944 | res = [] | |
1945 | for pagenum in itertools.count(start // self._pagesize): | |
1946 | firstid = pagenum * self._pagesize | |
1947 | nextfirstid = pagenum * self._pagesize + self._pagesize | |
1948 | if start >= nextfirstid: | |
1949 | continue | |
1950 | ||
b95dc034 YCH |
1951 | page_results = None |
1952 | if self._use_cache: | |
1953 | page_results = self._cache.get(pagenum) | |
1954 | if page_results is None: | |
1955 | page_results = list(self._pagefunc(pagenum)) | |
1956 | if self._use_cache: | |
1957 | self._cache[pagenum] = page_results | |
b7ab0590 PH |
1958 | |
1959 | startv = ( | |
1960 | start % self._pagesize | |
1961 | if firstid <= start < nextfirstid | |
1962 | else 0) | |
1963 | ||
1964 | endv = ( | |
1965 | ((end - 1) % self._pagesize) + 1 | |
1966 | if (end is not None and firstid <= end <= nextfirstid) | |
1967 | else None) | |
1968 | ||
1969 | if startv != 0 or endv is not None: | |
1970 | page_results = page_results[startv:endv] | |
1971 | res.extend(page_results) | |
1972 | ||
1973 | # A little optimization - if current page is not "full", ie. does | |
1974 | # not contain page_size videos then we can assume that this page | |
1975 | # is the last one - there are no more ids on further pages - | |
1976 | # i.e. no need to query again. | |
1977 | if len(page_results) + startv < self._pagesize: | |
1978 | break | |
1979 | ||
1980 | # If we got the whole page, but the next page is not interesting, | |
1981 | # break out early as well | |
1982 | if end == nextfirstid: | |
1983 | break | |
1984 | return res | |
81c2f20b PH |
1985 | |
1986 | ||
9c44d242 PH |
1987 | class InAdvancePagedList(PagedList): |
1988 | def __init__(self, pagefunc, pagecount, pagesize): | |
1989 | self._pagefunc = pagefunc | |
1990 | self._pagecount = pagecount | |
1991 | self._pagesize = pagesize | |
1992 | ||
1993 | def getslice(self, start=0, end=None): | |
1994 | res = [] | |
1995 | start_page = start // self._pagesize | |
1996 | end_page = ( | |
1997 | self._pagecount if end is None else (end // self._pagesize + 1)) | |
1998 | skip_elems = start - start_page * self._pagesize | |
1999 | only_more = None if end is None else end - start | |
2000 | for pagenum in range(start_page, end_page): | |
2001 | page = list(self._pagefunc(pagenum)) | |
2002 | if skip_elems: | |
2003 | page = page[skip_elems:] | |
2004 | skip_elems = None | |
2005 | if only_more is not None: | |
2006 | if len(page) < only_more: | |
2007 | only_more -= len(page) | |
2008 | else: | |
2009 | page = page[:only_more] | |
2010 | res.extend(page) | |
2011 | break | |
2012 | res.extend(page) | |
2013 | return res | |
2014 | ||
2015 | ||
81c2f20b | 2016 | def uppercase_escape(s): |
676eb3f2 | 2017 | unicode_escape = codecs.getdecoder('unicode_escape') |
81c2f20b | 2018 | return re.sub( |
a612753d | 2019 | r'\\U[0-9a-fA-F]{8}', |
676eb3f2 PH |
2020 | lambda m: unicode_escape(m.group(0))[0], |
2021 | s) | |
0fe2ff78 YCH |
2022 | |
2023 | ||
2024 | def lowercase_escape(s): | |
2025 | unicode_escape = codecs.getdecoder('unicode_escape') | |
2026 | return re.sub( | |
2027 | r'\\u[0-9a-fA-F]{4}', | |
2028 | lambda m: unicode_escape(m.group(0))[0], | |
2029 | s) | |
b53466e1 | 2030 | |
d05cfe06 S |
2031 | |
2032 | def escape_rfc3986(s): | |
2033 | """Escape non-ASCII characters as suggested by RFC 3986""" | |
8f9312c3 | 2034 | if sys.version_info < (3, 0) and isinstance(s, compat_str): |
d05cfe06 | 2035 | s = s.encode('utf-8') |
ecc0c5ee | 2036 | return compat_urllib_parse.quote(s, b"%/;:@&=+$,!~*'()?#[]") |
d05cfe06 S |
2037 | |
2038 | ||
2039 | def escape_url(url): | |
2040 | """Escape URL as suggested by RFC 3986""" | |
2041 | url_parsed = compat_urllib_parse_urlparse(url) | |
2042 | return url_parsed._replace( | |
efbed08d | 2043 | netloc=url_parsed.netloc.encode('idna').decode('ascii'), |
d05cfe06 S |
2044 | path=escape_rfc3986(url_parsed.path), |
2045 | params=escape_rfc3986(url_parsed.params), | |
2046 | query=escape_rfc3986(url_parsed.query), | |
2047 | fragment=escape_rfc3986(url_parsed.fragment) | |
2048 | ).geturl() | |
2049 | ||
62e609ab PH |
2050 | |
2051 | def read_batch_urls(batch_fd): | |
2052 | def fixup(url): | |
2053 | if not isinstance(url, compat_str): | |
2054 | url = url.decode('utf-8', 'replace') | |
28e614de | 2055 | BOM_UTF8 = '\xef\xbb\xbf' |
62e609ab PH |
2056 | if url.startswith(BOM_UTF8): |
2057 | url = url[len(BOM_UTF8):] | |
2058 | url = url.strip() | |
2059 | if url.startswith(('#', ';', ']')): | |
2060 | return False | |
2061 | return url | |
2062 | ||
2063 | with contextlib.closing(batch_fd) as fd: | |
2064 | return [url for url in map(fixup, fd) if url] | |
b74fa8cd JMF |
2065 | |
2066 | ||
2067 | def urlencode_postdata(*args, **kargs): | |
15707c7e | 2068 | return compat_urllib_parse_urlencode(*args, **kargs).encode('ascii') |
bcf89ce6 PH |
2069 | |
2070 | ||
38f9ef31 | 2071 | def update_url_query(url, query): |
cacd9966 YCH |
2072 | if not query: |
2073 | return url | |
38f9ef31 | 2074 | parsed_url = compat_urlparse.urlparse(url) |
2075 | qs = compat_parse_qs(parsed_url.query) | |
2076 | qs.update(query) | |
2077 | return compat_urlparse.urlunparse(parsed_url._replace( | |
15707c7e | 2078 | query=compat_urllib_parse_urlencode(qs, True))) |
16392824 | 2079 | |
8e60dc75 | 2080 | |
ed0291d1 S |
2081 | def update_Request(req, url=None, data=None, headers={}, query={}): |
2082 | req_headers = req.headers.copy() | |
2083 | req_headers.update(headers) | |
2084 | req_data = data or req.data | |
2085 | req_url = update_url_query(url or req.get_full_url(), query) | |
95cf60e8 S |
2086 | req_get_method = req.get_method() |
2087 | if req_get_method == 'HEAD': | |
2088 | req_type = HEADRequest | |
2089 | elif req_get_method == 'PUT': | |
2090 | req_type = PUTRequest | |
2091 | else: | |
2092 | req_type = compat_urllib_request.Request | |
ed0291d1 S |
2093 | new_req = req_type( |
2094 | req_url, data=req_data, headers=req_headers, | |
2095 | origin_req_host=req.origin_req_host, unverifiable=req.unverifiable) | |
2096 | if hasattr(req, 'timeout'): | |
2097 | new_req.timeout = req.timeout | |
2098 | return new_req | |
2099 | ||
2100 | ||
10c87c15 | 2101 | def _multipart_encode_impl(data, boundary): |
0c265486 YCH |
2102 | content_type = 'multipart/form-data; boundary=%s' % boundary |
2103 | ||
2104 | out = b'' | |
2105 | for k, v in data.items(): | |
2106 | out += b'--' + boundary.encode('ascii') + b'\r\n' | |
2107 | if isinstance(k, compat_str): | |
2108 | k = k.encode('utf-8') | |
2109 | if isinstance(v, compat_str): | |
2110 | v = v.encode('utf-8') | |
2111 | # RFC 2047 requires non-ASCII field names to be encoded, while RFC 7578 | |
2112 | # suggests sending UTF-8 directly. Firefox sends UTF-8, too | |
b2ad479d | 2113 | content = b'Content-Disposition: form-data; name="' + k + b'"\r\n\r\n' + v + b'\r\n' |
0c265486 YCH |
2114 | if boundary.encode('ascii') in content: |
2115 | raise ValueError('Boundary overlaps with data') | |
2116 | out += content | |
2117 | ||
2118 | out += b'--' + boundary.encode('ascii') + b'--\r\n' | |
2119 | ||
2120 | return out, content_type | |
2121 | ||
2122 | ||
2123 | def multipart_encode(data, boundary=None): | |
2124 | ''' | |
2125 | Encode a dict to RFC 7578-compliant form-data | |
2126 | ||
2127 | data: | |
2128 | A dict where keys and values can be either Unicode or bytes-like | |
2129 | objects. | |
2130 | boundary: | |
2131 | If specified a Unicode object, it's used as the boundary. Otherwise | |
2132 | a random boundary is generated. | |
2133 | ||
2134 | Reference: https://tools.ietf.org/html/rfc7578 | |
2135 | ''' | |
2136 | has_specified_boundary = boundary is not None | |
2137 | ||
2138 | while True: | |
2139 | if boundary is None: | |
2140 | boundary = '---------------' + str(random.randrange(0x0fffffff, 0xffffffff)) | |
2141 | ||
2142 | try: | |
10c87c15 | 2143 | out, content_type = _multipart_encode_impl(data, boundary) |
0c265486 YCH |
2144 | break |
2145 | except ValueError: | |
2146 | if has_specified_boundary: | |
2147 | raise | |
2148 | boundary = None | |
2149 | ||
2150 | return out, content_type | |
2151 | ||
2152 | ||
86296ad2 | 2153 | def dict_get(d, key_or_keys, default=None, skip_false_values=True): |
cbecc9b9 S |
2154 | if isinstance(key_or_keys, (list, tuple)): |
2155 | for key in key_or_keys: | |
86296ad2 S |
2156 | if key not in d or d[key] is None or skip_false_values and not d[key]: |
2157 | continue | |
2158 | return d[key] | |
cbecc9b9 S |
2159 | return default |
2160 | return d.get(key_or_keys, default) | |
2161 | ||
2162 | ||
329ca3be | 2163 | def try_get(src, getter, expected_type=None): |
a32a9a7e S |
2164 | if not isinstance(getter, (list, tuple)): |
2165 | getter = [getter] | |
2166 | for get in getter: | |
2167 | try: | |
2168 | v = get(src) | |
2169 | except (AttributeError, KeyError, TypeError, IndexError): | |
2170 | pass | |
2171 | else: | |
2172 | if expected_type is None or isinstance(v, expected_type): | |
2173 | return v | |
329ca3be S |
2174 | |
2175 | ||
8e60dc75 S |
2176 | def encode_compat_str(string, encoding=preferredencoding(), errors='strict'): |
2177 | return string if isinstance(string, compat_str) else compat_str(string, encoding, errors) | |
2178 | ||
16392824 | 2179 | |
a1a530b0 PH |
2180 | US_RATINGS = { |
2181 | 'G': 0, | |
2182 | 'PG': 10, | |
2183 | 'PG-13': 13, | |
2184 | 'R': 16, | |
2185 | 'NC': 18, | |
2186 | } | |
fac55558 PH |
2187 | |
2188 | ||
a8795327 S |
2189 | TV_PARENTAL_GUIDELINES = { |
2190 | 'TV-Y': 0, | |
2191 | 'TV-Y7': 7, | |
2192 | 'TV-G': 0, | |
2193 | 'TV-PG': 0, | |
2194 | 'TV-14': 14, | |
2195 | 'TV-MA': 17, | |
2196 | } | |
2197 | ||
2198 | ||
146c80e2 | 2199 | def parse_age_limit(s): |
a8795327 S |
2200 | if type(s) == int: |
2201 | return s if 0 <= s <= 21 else None | |
2202 | if not isinstance(s, compat_basestring): | |
d838b1bd | 2203 | return None |
146c80e2 | 2204 | m = re.match(r'^(?P<age>\d{1,2})\+?$', s) |
a8795327 S |
2205 | if m: |
2206 | return int(m.group('age')) | |
2207 | if s in US_RATINGS: | |
2208 | return US_RATINGS[s] | |
2209 | return TV_PARENTAL_GUIDELINES.get(s) | |
146c80e2 S |
2210 | |
2211 | ||
fac55558 | 2212 | def strip_jsonp(code): |
609a61e3 | 2213 | return re.sub( |
5552c9eb YCH |
2214 | r'''(?sx)^ |
2215 | (?:window\.)?(?P<func_name>[a-zA-Z0-9_.$]+) | |
2216 | (?:\s*&&\s*(?P=func_name))? | |
2217 | \s*\(\s*(?P<callback_data>.*)\);? | |
2218 | \s*?(?://[^\n]*)*$''', | |
2219 | r'\g<callback_data>', code) | |
478c2c61 PH |
2220 | |
2221 | ||
e05f6939 | 2222 | def js_to_json(code): |
4195096e S |
2223 | COMMENT_RE = r'/\*(?:(?!\*/).)*?\*/|//[^\n]*' |
2224 | SKIP_RE = r'\s*(?:{comment})?\s*'.format(comment=COMMENT_RE) | |
2225 | INTEGER_TABLE = ( | |
2226 | (r'(?s)^(0[xX][0-9a-fA-F]+){skip}:?$'.format(skip=SKIP_RE), 16), | |
2227 | (r'(?s)^(0+[0-7]+){skip}:?$'.format(skip=SKIP_RE), 8), | |
2228 | ) | |
2229 | ||
e05f6939 | 2230 | def fix_kv(m): |
e7b6d122 PH |
2231 | v = m.group(0) |
2232 | if v in ('true', 'false', 'null'): | |
2233 | return v | |
b3ee552e | 2234 | elif v.startswith('/*') or v.startswith('//') or v == ',': |
bd1e4844 | 2235 | return "" |
2236 | ||
2237 | if v[0] in ("'", '"'): | |
2238 | v = re.sub(r'(?s)\\.|"', lambda m: { | |
e7b6d122 | 2239 | '"': '\\"', |
bd1e4844 | 2240 | "\\'": "'", |
2241 | '\\\n': '', | |
2242 | '\\x': '\\u00', | |
2243 | }.get(m.group(0), m.group(0)), v[1:-1]) | |
2244 | ||
89ac4a19 S |
2245 | for regex, base in INTEGER_TABLE: |
2246 | im = re.match(regex, v) | |
2247 | if im: | |
e4659b45 | 2248 | i = int(im.group(1), base) |
89ac4a19 S |
2249 | return '"%d":' % i if v.endswith(':') else '%d' % i |
2250 | ||
e7b6d122 | 2251 | return '"%s"' % v |
e05f6939 | 2252 | |
bd1e4844 | 2253 | return re.sub(r'''(?sx) |
2254 | "(?:[^"\\]*(?:\\\\|\\['"nurtbfx/\n]))*[^"\\]*"| | |
2255 | '(?:[^'\\]*(?:\\\\|\\['"nurtbfx/\n]))*[^'\\]*'| | |
4195096e | 2256 | {comment}|,(?={skip}[\]}}])| |
bd1e4844 | 2257 | [a-zA-Z_][.a-zA-Z_0-9]*| |
4195096e S |
2258 | \b(?:0[xX][0-9a-fA-F]+|0+[0-7]+)(?:{skip}:)?| |
2259 | [0-9]+(?={skip}:) | |
2260 | '''.format(comment=COMMENT_RE, skip=SKIP_RE), fix_kv, code) | |
e05f6939 PH |
2261 | |
2262 | ||
478c2c61 PH |
2263 | def qualities(quality_ids): |
2264 | """ Get a numeric quality value out of a list of possible values """ | |
2265 | def q(qid): | |
2266 | try: | |
2267 | return quality_ids.index(qid) | |
2268 | except ValueError: | |
2269 | return -1 | |
2270 | return q | |
2271 | ||
acd69589 PH |
2272 | |
2273 | DEFAULT_OUTTMPL = '%(title)s-%(id)s.%(ext)s' | |
0a871f68 | 2274 | |
a020a0dc PH |
2275 | |
2276 | def limit_length(s, length): | |
2277 | """ Add ellipses to overly long strings """ | |
2278 | if s is None: | |
2279 | return None | |
2280 | ELLIPSES = '...' | |
2281 | if len(s) > length: | |
2282 | return s[:length - len(ELLIPSES)] + ELLIPSES | |
2283 | return s | |
48844745 PH |
2284 | |
2285 | ||
2286 | def version_tuple(v): | |
5f9b8394 | 2287 | return tuple(int(e) for e in re.split(r'[-.]', v)) |
48844745 PH |
2288 | |
2289 | ||
2290 | def is_outdated_version(version, limit, assume_new=True): | |
2291 | if not version: | |
2292 | return not assume_new | |
2293 | try: | |
2294 | return version_tuple(version) < version_tuple(limit) | |
2295 | except ValueError: | |
2296 | return not assume_new | |
732ea2f0 PH |
2297 | |
2298 | ||
2299 | def ytdl_is_updateable(): | |
2300 | """ Returns if youtube-dl can be updated with -U """ | |
2301 | from zipimport import zipimporter | |
2302 | ||
2303 | return isinstance(globals().get('__loader__'), zipimporter) or hasattr(sys, 'frozen') | |
7d4111ed PH |
2304 | |
2305 | ||
2306 | def args_to_str(args): | |
2307 | # Get a short string representation for a subprocess command | |
702ccf2d | 2308 | return ' '.join(compat_shlex_quote(a) for a in args) |
2ccd1b10 PH |
2309 | |
2310 | ||
9b9c5355 | 2311 | def error_to_compat_str(err): |
fdae2358 S |
2312 | err_str = str(err) |
2313 | # On python 2 error byte string must be decoded with proper | |
2314 | # encoding rather than ascii | |
2315 | if sys.version_info[0] < 3: | |
2316 | err_str = err_str.decode(preferredencoding()) | |
2317 | return err_str | |
2318 | ||
2319 | ||
c460bdd5 | 2320 | def mimetype2ext(mt): |
eb9ee194 S |
2321 | if mt is None: |
2322 | return None | |
2323 | ||
765ac263 JMF |
2324 | ext = { |
2325 | 'audio/mp4': 'm4a', | |
6c33d24b YCH |
2326 | # Per RFC 3003, audio/mpeg can be .mp1, .mp2 or .mp3. Here use .mp3 as |
2327 | # it's the most popular one | |
2328 | 'audio/mpeg': 'mp3', | |
765ac263 JMF |
2329 | }.get(mt) |
2330 | if ext is not None: | |
2331 | return ext | |
2332 | ||
c460bdd5 | 2333 | _, _, res = mt.rpartition('/') |
6562d34a | 2334 | res = res.split(';')[0].strip().lower() |
c460bdd5 PH |
2335 | |
2336 | return { | |
f6861ec9 | 2337 | '3gpp': '3gp', |
cafcf657 | 2338 | 'smptett+xml': 'tt', |
cafcf657 | 2339 | 'ttaf+xml': 'dfxp', |
a0d8d704 | 2340 | 'ttml+xml': 'ttml', |
f6861ec9 | 2341 | 'x-flv': 'flv', |
a0d8d704 YCH |
2342 | 'x-mp4-fragmented': 'mp4', |
2343 | 'x-ms-wmv': 'wmv', | |
b4173f15 RA |
2344 | 'mpegurl': 'm3u8', |
2345 | 'x-mpegurl': 'm3u8', | |
2346 | 'vnd.apple.mpegurl': 'm3u8', | |
2347 | 'dash+xml': 'mpd', | |
b4173f15 | 2348 | 'f4m+xml': 'f4m', |
f164b971 | 2349 | 'hds+xml': 'f4m', |
e910fe2f | 2350 | 'vnd.ms-sstr+xml': 'ism', |
c2b2c7e1 | 2351 | 'quicktime': 'mov', |
98ce1a3f | 2352 | 'mp2t': 'ts', |
c460bdd5 PH |
2353 | }.get(res, res) |
2354 | ||
2355 | ||
4f3c5e06 | 2356 | def parse_codecs(codecs_str): |
2357 | # http://tools.ietf.org/html/rfc6381 | |
2358 | if not codecs_str: | |
2359 | return {} | |
2360 | splited_codecs = list(filter(None, map( | |
2361 | lambda str: str.strip(), codecs_str.strip().strip(',').split(',')))) | |
2362 | vcodec, acodec = None, None | |
2363 | for full_codec in splited_codecs: | |
2364 | codec = full_codec.split('.')[0] | |
2365 | if codec in ('avc1', 'avc2', 'avc3', 'avc4', 'vp9', 'vp8', 'hev1', 'hev2', 'h263', 'h264', 'mp4v'): | |
2366 | if not vcodec: | |
2367 | vcodec = full_codec | |
60f5c9fb | 2368 | elif codec in ('mp4a', 'opus', 'vorbis', 'mp3', 'aac', 'ac-3', 'ec-3', 'eac3', 'dtsc', 'dtse', 'dtsh', 'dtsl'): |
4f3c5e06 | 2369 | if not acodec: |
2370 | acodec = full_codec | |
2371 | else: | |
60f5c9fb | 2372 | write_string('WARNING: Unknown codec %s\n' % full_codec, sys.stderr) |
4f3c5e06 | 2373 | if not vcodec and not acodec: |
2374 | if len(splited_codecs) == 2: | |
2375 | return { | |
2376 | 'vcodec': vcodec, | |
2377 | 'acodec': acodec, | |
2378 | } | |
2379 | elif len(splited_codecs) == 1: | |
2380 | return { | |
2381 | 'vcodec': 'none', | |
2382 | 'acodec': vcodec, | |
2383 | } | |
2384 | else: | |
2385 | return { | |
2386 | 'vcodec': vcodec or 'none', | |
2387 | 'acodec': acodec or 'none', | |
2388 | } | |
2389 | return {} | |
2390 | ||
2391 | ||
2ccd1b10 | 2392 | def urlhandle_detect_ext(url_handle): |
79298173 | 2393 | getheader = url_handle.headers.get |
2ccd1b10 | 2394 | |
b55ee18f PH |
2395 | cd = getheader('Content-Disposition') |
2396 | if cd: | |
2397 | m = re.match(r'attachment;\s*filename="(?P<filename>[^"]+)"', cd) | |
2398 | if m: | |
2399 | e = determine_ext(m.group('filename'), default_ext=None) | |
2400 | if e: | |
2401 | return e | |
2402 | ||
c460bdd5 | 2403 | return mimetype2ext(getheader('Content-Type')) |
05900629 PH |
2404 | |
2405 | ||
1e399778 YCH |
2406 | def encode_data_uri(data, mime_type): |
2407 | return 'data:%s;base64,%s' % (mime_type, base64.b64encode(data).decode('ascii')) | |
2408 | ||
2409 | ||
05900629 | 2410 | def age_restricted(content_limit, age_limit): |
6ec6cb4e | 2411 | """ Returns True iff the content should be blocked """ |
05900629 PH |
2412 | |
2413 | if age_limit is None: # No limit set | |
2414 | return False | |
2415 | if content_limit is None: | |
2416 | return False # Content available for everyone | |
2417 | return age_limit < content_limit | |
61ca9a80 PH |
2418 | |
2419 | ||
2420 | def is_html(first_bytes): | |
2421 | """ Detect whether a file contains HTML by examining its first bytes. """ | |
2422 | ||
2423 | BOMS = [ | |
2424 | (b'\xef\xbb\xbf', 'utf-8'), | |
2425 | (b'\x00\x00\xfe\xff', 'utf-32-be'), | |
2426 | (b'\xff\xfe\x00\x00', 'utf-32-le'), | |
2427 | (b'\xff\xfe', 'utf-16-le'), | |
2428 | (b'\xfe\xff', 'utf-16-be'), | |
2429 | ] | |
2430 | for bom, enc in BOMS: | |
2431 | if first_bytes.startswith(bom): | |
2432 | s = first_bytes[len(bom):].decode(enc, 'replace') | |
2433 | break | |
2434 | else: | |
2435 | s = first_bytes.decode('utf-8', 'replace') | |
2436 | ||
2437 | return re.match(r'^\s*<', s) | |
a055469f PH |
2438 | |
2439 | ||
2440 | def determine_protocol(info_dict): | |
2441 | protocol = info_dict.get('protocol') | |
2442 | if protocol is not None: | |
2443 | return protocol | |
2444 | ||
2445 | url = info_dict['url'] | |
2446 | if url.startswith('rtmp'): | |
2447 | return 'rtmp' | |
2448 | elif url.startswith('mms'): | |
2449 | return 'mms' | |
2450 | elif url.startswith('rtsp'): | |
2451 | return 'rtsp' | |
2452 | ||
2453 | ext = determine_ext(url) | |
2454 | if ext == 'm3u8': | |
2455 | return 'm3u8' | |
2456 | elif ext == 'f4m': | |
2457 | return 'f4m' | |
2458 | ||
2459 | return compat_urllib_parse_urlparse(url).scheme | |
cfb56d1a PH |
2460 | |
2461 | ||
2462 | def render_table(header_row, data): | |
2463 | """ Render a list of rows, each as a list of values """ | |
2464 | table = [header_row] + data | |
2465 | max_lens = [max(len(compat_str(v)) for v in col) for col in zip(*table)] | |
2466 | format_str = ' '.join('%-' + compat_str(ml + 1) + 's' for ml in max_lens[:-1]) + '%s' | |
2467 | return '\n'.join(format_str % tuple(row) for row in table) | |
347de493 PH |
2468 | |
2469 | ||
2470 | def _match_one(filter_part, dct): | |
2471 | COMPARISON_OPERATORS = { | |
2472 | '<': operator.lt, | |
2473 | '<=': operator.le, | |
2474 | '>': operator.gt, | |
2475 | '>=': operator.ge, | |
2476 | '=': operator.eq, | |
2477 | '!=': operator.ne, | |
2478 | } | |
2479 | operator_rex = re.compile(r'''(?x)\s* | |
2480 | (?P<key>[a-z_]+) | |
2481 | \s*(?P<op>%s)(?P<none_inclusive>\s*\?)?\s* | |
2482 | (?: | |
2483 | (?P<intval>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)| | |
db13c16e | 2484 | (?P<quote>["\'])(?P<quotedstrval>(?:\\.|(?!(?P=quote)|\\).)+?)(?P=quote)| |
347de493 PH |
2485 | (?P<strval>(?![0-9.])[a-z0-9A-Z]*) |
2486 | ) | |
2487 | \s*$ | |
2488 | ''' % '|'.join(map(re.escape, COMPARISON_OPERATORS.keys()))) | |
2489 | m = operator_rex.search(filter_part) | |
2490 | if m: | |
2491 | op = COMPARISON_OPERATORS[m.group('op')] | |
e5a088dc | 2492 | actual_value = dct.get(m.group('key')) |
db13c16e S |
2493 | if (m.group('quotedstrval') is not None or |
2494 | m.group('strval') is not None or | |
e5a088dc S |
2495 | # If the original field is a string and matching comparisonvalue is |
2496 | # a number we should respect the origin of the original field | |
2497 | # and process comparison value as a string (see | |
2498 | # https://github.com/rg3/youtube-dl/issues/11082). | |
2499 | actual_value is not None and m.group('intval') is not None and | |
2500 | isinstance(actual_value, compat_str)): | |
347de493 PH |
2501 | if m.group('op') not in ('=', '!='): |
2502 | raise ValueError( | |
2503 | 'Operator %s does not support string values!' % m.group('op')) | |
db13c16e S |
2504 | comparison_value = m.group('quotedstrval') or m.group('strval') or m.group('intval') |
2505 | quote = m.group('quote') | |
2506 | if quote is not None: | |
2507 | comparison_value = comparison_value.replace(r'\%s' % quote, quote) | |
347de493 PH |
2508 | else: |
2509 | try: | |
2510 | comparison_value = int(m.group('intval')) | |
2511 | except ValueError: | |
2512 | comparison_value = parse_filesize(m.group('intval')) | |
2513 | if comparison_value is None: | |
2514 | comparison_value = parse_filesize(m.group('intval') + 'B') | |
2515 | if comparison_value is None: | |
2516 | raise ValueError( | |
2517 | 'Invalid integer value %r in filter part %r' % ( | |
2518 | m.group('intval'), filter_part)) | |
347de493 PH |
2519 | if actual_value is None: |
2520 | return m.group('none_inclusive') | |
2521 | return op(actual_value, comparison_value) | |
2522 | ||
2523 | UNARY_OPERATORS = { | |
2524 | '': lambda v: v is not None, | |
2525 | '!': lambda v: v is None, | |
2526 | } | |
2527 | operator_rex = re.compile(r'''(?x)\s* | |
2528 | (?P<op>%s)\s*(?P<key>[a-z_]+) | |
2529 | \s*$ | |
2530 | ''' % '|'.join(map(re.escape, UNARY_OPERATORS.keys()))) | |
2531 | m = operator_rex.search(filter_part) | |
2532 | if m: | |
2533 | op = UNARY_OPERATORS[m.group('op')] | |
2534 | actual_value = dct.get(m.group('key')) | |
2535 | return op(actual_value) | |
2536 | ||
2537 | raise ValueError('Invalid filter part %r' % filter_part) | |
2538 | ||
2539 | ||
2540 | def match_str(filter_str, dct): | |
2541 | """ Filter a dictionary with a simple string syntax. Returns True (=passes filter) or false """ | |
2542 | ||
2543 | return all( | |
2544 | _match_one(filter_part, dct) for filter_part in filter_str.split('&')) | |
2545 | ||
2546 | ||
2547 | def match_filter_func(filter_str): | |
2548 | def _match_func(info_dict): | |
2549 | if match_str(filter_str, info_dict): | |
2550 | return None | |
2551 | else: | |
2552 | video_title = info_dict.get('title', info_dict.get('id', 'video')) | |
2553 | return '%s does not pass filter %s, skipping ..' % (video_title, filter_str) | |
2554 | return _match_func | |
91410c9b PH |
2555 | |
2556 | ||
bf6427d2 YCH |
2557 | def parse_dfxp_time_expr(time_expr): |
2558 | if not time_expr: | |
d631d5f9 | 2559 | return |
bf6427d2 YCH |
2560 | |
2561 | mobj = re.match(r'^(?P<time_offset>\d+(?:\.\d+)?)s?$', time_expr) | |
2562 | if mobj: | |
2563 | return float(mobj.group('time_offset')) | |
2564 | ||
db2fe38b | 2565 | mobj = re.match(r'^(\d+):(\d\d):(\d\d(?:(?:\.|:)\d+)?)$', time_expr) |
bf6427d2 | 2566 | if mobj: |
db2fe38b | 2567 | return 3600 * int(mobj.group(1)) + 60 * int(mobj.group(2)) + float(mobj.group(3).replace(':', '.')) |
bf6427d2 YCH |
2568 | |
2569 | ||
c1c924ab YCH |
2570 | def srt_subtitles_timecode(seconds): |
2571 | return '%02d:%02d:%02d,%03d' % (seconds / 3600, (seconds % 3600) / 60, seconds % 60, (seconds % 1) * 1000) | |
bf6427d2 YCH |
2572 | |
2573 | ||
2574 | def dfxp2srt(dfxp_data): | |
3869028f YCH |
2575 | ''' |
2576 | @param dfxp_data A bytes-like object containing DFXP data | |
2577 | @returns A unicode object containing converted SRT data | |
2578 | ''' | |
5b995f71 | 2579 | LEGACY_NAMESPACES = ( |
3869028f YCH |
2580 | (b'http://www.w3.org/ns/ttml', [ |
2581 | b'http://www.w3.org/2004/11/ttaf1', | |
2582 | b'http://www.w3.org/2006/04/ttaf1', | |
2583 | b'http://www.w3.org/2006/10/ttaf1', | |
5b995f71 | 2584 | ]), |
3869028f YCH |
2585 | (b'http://www.w3.org/ns/ttml#styling', [ |
2586 | b'http://www.w3.org/ns/ttml#style', | |
5b995f71 RA |
2587 | ]), |
2588 | ) | |
2589 | ||
2590 | SUPPORTED_STYLING = [ | |
2591 | 'color', | |
2592 | 'fontFamily', | |
2593 | 'fontSize', | |
2594 | 'fontStyle', | |
2595 | 'fontWeight', | |
2596 | 'textDecoration' | |
2597 | ] | |
2598 | ||
4e335771 YCH |
2599 | _x = functools.partial(xpath_with_ns, ns_map={ |
2600 | 'ttml': 'http://www.w3.org/ns/ttml', | |
5b995f71 | 2601 | 'tts': 'http://www.w3.org/ns/ttml#styling', |
4e335771 | 2602 | }) |
bf6427d2 | 2603 | |
5b995f71 RA |
2604 | styles = {} |
2605 | default_style = {} | |
2606 | ||
87de7069 | 2607 | class TTMLPElementParser(object): |
5b995f71 RA |
2608 | _out = '' |
2609 | _unclosed_elements = [] | |
2610 | _applied_styles = [] | |
bf6427d2 | 2611 | |
2b14cb56 | 2612 | def start(self, tag, attrib): |
5b995f71 RA |
2613 | if tag in (_x('ttml:br'), 'br'): |
2614 | self._out += '\n' | |
2615 | else: | |
2616 | unclosed_elements = [] | |
2617 | style = {} | |
2618 | element_style_id = attrib.get('style') | |
2619 | if default_style: | |
2620 | style.update(default_style) | |
2621 | if element_style_id: | |
2622 | style.update(styles.get(element_style_id, {})) | |
2623 | for prop in SUPPORTED_STYLING: | |
2624 | prop_val = attrib.get(_x('tts:' + prop)) | |
2625 | if prop_val: | |
2626 | style[prop] = prop_val | |
2627 | if style: | |
2628 | font = '' | |
2629 | for k, v in sorted(style.items()): | |
2630 | if self._applied_styles and self._applied_styles[-1].get(k) == v: | |
2631 | continue | |
2632 | if k == 'color': | |
2633 | font += ' color="%s"' % v | |
2634 | elif k == 'fontSize': | |
2635 | font += ' size="%s"' % v | |
2636 | elif k == 'fontFamily': | |
2637 | font += ' face="%s"' % v | |
2638 | elif k == 'fontWeight' and v == 'bold': | |
2639 | self._out += '<b>' | |
2640 | unclosed_elements.append('b') | |
2641 | elif k == 'fontStyle' and v == 'italic': | |
2642 | self._out += '<i>' | |
2643 | unclosed_elements.append('i') | |
2644 | elif k == 'textDecoration' and v == 'underline': | |
2645 | self._out += '<u>' | |
2646 | unclosed_elements.append('u') | |
2647 | if font: | |
2648 | self._out += '<font' + font + '>' | |
2649 | unclosed_elements.append('font') | |
2650 | applied_style = {} | |
2651 | if self._applied_styles: | |
2652 | applied_style.update(self._applied_styles[-1]) | |
2653 | applied_style.update(style) | |
2654 | self._applied_styles.append(applied_style) | |
2655 | self._unclosed_elements.append(unclosed_elements) | |
bf6427d2 | 2656 | |
2b14cb56 | 2657 | def end(self, tag): |
5b995f71 RA |
2658 | if tag not in (_x('ttml:br'), 'br'): |
2659 | unclosed_elements = self._unclosed_elements.pop() | |
2660 | for element in reversed(unclosed_elements): | |
2661 | self._out += '</%s>' % element | |
2662 | if unclosed_elements and self._applied_styles: | |
2663 | self._applied_styles.pop() | |
bf6427d2 | 2664 | |
2b14cb56 | 2665 | def data(self, data): |
5b995f71 | 2666 | self._out += data |
2b14cb56 | 2667 | |
2668 | def close(self): | |
5b995f71 | 2669 | return self._out.strip() |
2b14cb56 | 2670 | |
2671 | def parse_node(node): | |
2672 | target = TTMLPElementParser() | |
2673 | parser = xml.etree.ElementTree.XMLParser(target=target) | |
2674 | parser.feed(xml.etree.ElementTree.tostring(node)) | |
2675 | return parser.close() | |
bf6427d2 | 2676 | |
5b995f71 RA |
2677 | for k, v in LEGACY_NAMESPACES: |
2678 | for ns in v: | |
2679 | dfxp_data = dfxp_data.replace(ns, k) | |
2680 | ||
3869028f | 2681 | dfxp = compat_etree_fromstring(dfxp_data) |
bf6427d2 | 2682 | out = [] |
5b995f71 | 2683 | paras = dfxp.findall(_x('.//ttml:p')) or dfxp.findall('.//p') |
1b0427e6 YCH |
2684 | |
2685 | if not paras: | |
2686 | raise ValueError('Invalid dfxp/TTML subtitle') | |
bf6427d2 | 2687 | |
5b995f71 RA |
2688 | repeat = False |
2689 | while True: | |
2690 | for style in dfxp.findall(_x('.//ttml:style')): | |
2691 | style_id = style.get('id') | |
2692 | parent_style_id = style.get('style') | |
2693 | if parent_style_id: | |
2694 | if parent_style_id not in styles: | |
2695 | repeat = True | |
2696 | continue | |
2697 | styles[style_id] = styles[parent_style_id].copy() | |
2698 | for prop in SUPPORTED_STYLING: | |
2699 | prop_val = style.get(_x('tts:' + prop)) | |
2700 | if prop_val: | |
2701 | styles.setdefault(style_id, {})[prop] = prop_val | |
2702 | if repeat: | |
2703 | repeat = False | |
2704 | else: | |
2705 | break | |
2706 | ||
2707 | for p in ('body', 'div'): | |
2708 | ele = xpath_element(dfxp, [_x('.//ttml:' + p), './/' + p]) | |
2709 | if ele is None: | |
2710 | continue | |
2711 | style = styles.get(ele.get('style')) | |
2712 | if not style: | |
2713 | continue | |
2714 | default_style.update(style) | |
2715 | ||
bf6427d2 | 2716 | for para, index in zip(paras, itertools.count(1)): |
d631d5f9 | 2717 | begin_time = parse_dfxp_time_expr(para.attrib.get('begin')) |
7dff0363 | 2718 | end_time = parse_dfxp_time_expr(para.attrib.get('end')) |
d631d5f9 YCH |
2719 | dur = parse_dfxp_time_expr(para.attrib.get('dur')) |
2720 | if begin_time is None: | |
2721 | continue | |
7dff0363 | 2722 | if not end_time: |
d631d5f9 YCH |
2723 | if not dur: |
2724 | continue | |
2725 | end_time = begin_time + dur | |
bf6427d2 YCH |
2726 | out.append('%d\n%s --> %s\n%s\n\n' % ( |
2727 | index, | |
c1c924ab YCH |
2728 | srt_subtitles_timecode(begin_time), |
2729 | srt_subtitles_timecode(end_time), | |
bf6427d2 YCH |
2730 | parse_node(para))) |
2731 | ||
2732 | return ''.join(out) | |
2733 | ||
2734 | ||
66e289ba S |
2735 | def cli_option(params, command_option, param): |
2736 | param = params.get(param) | |
98e698f1 RA |
2737 | if param: |
2738 | param = compat_str(param) | |
66e289ba S |
2739 | return [command_option, param] if param is not None else [] |
2740 | ||
2741 | ||
2742 | def cli_bool_option(params, command_option, param, true_value='true', false_value='false', separator=None): | |
2743 | param = params.get(param) | |
5b232f46 S |
2744 | if param is None: |
2745 | return [] | |
66e289ba S |
2746 | assert isinstance(param, bool) |
2747 | if separator: | |
2748 | return [command_option + separator + (true_value if param else false_value)] | |
2749 | return [command_option, true_value if param else false_value] | |
2750 | ||
2751 | ||
2752 | def cli_valueless_option(params, command_option, param, expected_value=True): | |
2753 | param = params.get(param) | |
2754 | return [command_option] if param == expected_value else [] | |
2755 | ||
2756 | ||
2757 | def cli_configuration_args(params, param, default=[]): | |
2758 | ex_args = params.get(param) | |
2759 | if ex_args is None: | |
2760 | return default | |
2761 | assert isinstance(ex_args, list) | |
2762 | return ex_args | |
2763 | ||
2764 | ||
39672624 YCH |
2765 | class ISO639Utils(object): |
2766 | # See http://www.loc.gov/standards/iso639-2/ISO-639-2_utf-8.txt | |
2767 | _lang_map = { | |
2768 | 'aa': 'aar', | |
2769 | 'ab': 'abk', | |
2770 | 'ae': 'ave', | |
2771 | 'af': 'afr', | |
2772 | 'ak': 'aka', | |
2773 | 'am': 'amh', | |
2774 | 'an': 'arg', | |
2775 | 'ar': 'ara', | |
2776 | 'as': 'asm', | |
2777 | 'av': 'ava', | |
2778 | 'ay': 'aym', | |
2779 | 'az': 'aze', | |
2780 | 'ba': 'bak', | |
2781 | 'be': 'bel', | |
2782 | 'bg': 'bul', | |
2783 | 'bh': 'bih', | |
2784 | 'bi': 'bis', | |
2785 | 'bm': 'bam', | |
2786 | 'bn': 'ben', | |
2787 | 'bo': 'bod', | |
2788 | 'br': 'bre', | |
2789 | 'bs': 'bos', | |
2790 | 'ca': 'cat', | |
2791 | 'ce': 'che', | |
2792 | 'ch': 'cha', | |
2793 | 'co': 'cos', | |
2794 | 'cr': 'cre', | |
2795 | 'cs': 'ces', | |
2796 | 'cu': 'chu', | |
2797 | 'cv': 'chv', | |
2798 | 'cy': 'cym', | |
2799 | 'da': 'dan', | |
2800 | 'de': 'deu', | |
2801 | 'dv': 'div', | |
2802 | 'dz': 'dzo', | |
2803 | 'ee': 'ewe', | |
2804 | 'el': 'ell', | |
2805 | 'en': 'eng', | |
2806 | 'eo': 'epo', | |
2807 | 'es': 'spa', | |
2808 | 'et': 'est', | |
2809 | 'eu': 'eus', | |
2810 | 'fa': 'fas', | |
2811 | 'ff': 'ful', | |
2812 | 'fi': 'fin', | |
2813 | 'fj': 'fij', | |
2814 | 'fo': 'fao', | |
2815 | 'fr': 'fra', | |
2816 | 'fy': 'fry', | |
2817 | 'ga': 'gle', | |
2818 | 'gd': 'gla', | |
2819 | 'gl': 'glg', | |
2820 | 'gn': 'grn', | |
2821 | 'gu': 'guj', | |
2822 | 'gv': 'glv', | |
2823 | 'ha': 'hau', | |
2824 | 'he': 'heb', | |
2825 | 'hi': 'hin', | |
2826 | 'ho': 'hmo', | |
2827 | 'hr': 'hrv', | |
2828 | 'ht': 'hat', | |
2829 | 'hu': 'hun', | |
2830 | 'hy': 'hye', | |
2831 | 'hz': 'her', | |
2832 | 'ia': 'ina', | |
2833 | 'id': 'ind', | |
2834 | 'ie': 'ile', | |
2835 | 'ig': 'ibo', | |
2836 | 'ii': 'iii', | |
2837 | 'ik': 'ipk', | |
2838 | 'io': 'ido', | |
2839 | 'is': 'isl', | |
2840 | 'it': 'ita', | |
2841 | 'iu': 'iku', | |
2842 | 'ja': 'jpn', | |
2843 | 'jv': 'jav', | |
2844 | 'ka': 'kat', | |
2845 | 'kg': 'kon', | |
2846 | 'ki': 'kik', | |
2847 | 'kj': 'kua', | |
2848 | 'kk': 'kaz', | |
2849 | 'kl': 'kal', | |
2850 | 'km': 'khm', | |
2851 | 'kn': 'kan', | |
2852 | 'ko': 'kor', | |
2853 | 'kr': 'kau', | |
2854 | 'ks': 'kas', | |
2855 | 'ku': 'kur', | |
2856 | 'kv': 'kom', | |
2857 | 'kw': 'cor', | |
2858 | 'ky': 'kir', | |
2859 | 'la': 'lat', | |
2860 | 'lb': 'ltz', | |
2861 | 'lg': 'lug', | |
2862 | 'li': 'lim', | |
2863 | 'ln': 'lin', | |
2864 | 'lo': 'lao', | |
2865 | 'lt': 'lit', | |
2866 | 'lu': 'lub', | |
2867 | 'lv': 'lav', | |
2868 | 'mg': 'mlg', | |
2869 | 'mh': 'mah', | |
2870 | 'mi': 'mri', | |
2871 | 'mk': 'mkd', | |
2872 | 'ml': 'mal', | |
2873 | 'mn': 'mon', | |
2874 | 'mr': 'mar', | |
2875 | 'ms': 'msa', | |
2876 | 'mt': 'mlt', | |
2877 | 'my': 'mya', | |
2878 | 'na': 'nau', | |
2879 | 'nb': 'nob', | |
2880 | 'nd': 'nde', | |
2881 | 'ne': 'nep', | |
2882 | 'ng': 'ndo', | |
2883 | 'nl': 'nld', | |
2884 | 'nn': 'nno', | |
2885 | 'no': 'nor', | |
2886 | 'nr': 'nbl', | |
2887 | 'nv': 'nav', | |
2888 | 'ny': 'nya', | |
2889 | 'oc': 'oci', | |
2890 | 'oj': 'oji', | |
2891 | 'om': 'orm', | |
2892 | 'or': 'ori', | |
2893 | 'os': 'oss', | |
2894 | 'pa': 'pan', | |
2895 | 'pi': 'pli', | |
2896 | 'pl': 'pol', | |
2897 | 'ps': 'pus', | |
2898 | 'pt': 'por', | |
2899 | 'qu': 'que', | |
2900 | 'rm': 'roh', | |
2901 | 'rn': 'run', | |
2902 | 'ro': 'ron', | |
2903 | 'ru': 'rus', | |
2904 | 'rw': 'kin', | |
2905 | 'sa': 'san', | |
2906 | 'sc': 'srd', | |
2907 | 'sd': 'snd', | |
2908 | 'se': 'sme', | |
2909 | 'sg': 'sag', | |
2910 | 'si': 'sin', | |
2911 | 'sk': 'slk', | |
2912 | 'sl': 'slv', | |
2913 | 'sm': 'smo', | |
2914 | 'sn': 'sna', | |
2915 | 'so': 'som', | |
2916 | 'sq': 'sqi', | |
2917 | 'sr': 'srp', | |
2918 | 'ss': 'ssw', | |
2919 | 'st': 'sot', | |
2920 | 'su': 'sun', | |
2921 | 'sv': 'swe', | |
2922 | 'sw': 'swa', | |
2923 | 'ta': 'tam', | |
2924 | 'te': 'tel', | |
2925 | 'tg': 'tgk', | |
2926 | 'th': 'tha', | |
2927 | 'ti': 'tir', | |
2928 | 'tk': 'tuk', | |
2929 | 'tl': 'tgl', | |
2930 | 'tn': 'tsn', | |
2931 | 'to': 'ton', | |
2932 | 'tr': 'tur', | |
2933 | 'ts': 'tso', | |
2934 | 'tt': 'tat', | |
2935 | 'tw': 'twi', | |
2936 | 'ty': 'tah', | |
2937 | 'ug': 'uig', | |
2938 | 'uk': 'ukr', | |
2939 | 'ur': 'urd', | |
2940 | 'uz': 'uzb', | |
2941 | 've': 'ven', | |
2942 | 'vi': 'vie', | |
2943 | 'vo': 'vol', | |
2944 | 'wa': 'wln', | |
2945 | 'wo': 'wol', | |
2946 | 'xh': 'xho', | |
2947 | 'yi': 'yid', | |
2948 | 'yo': 'yor', | |
2949 | 'za': 'zha', | |
2950 | 'zh': 'zho', | |
2951 | 'zu': 'zul', | |
2952 | } | |
2953 | ||
2954 | @classmethod | |
2955 | def short2long(cls, code): | |
2956 | """Convert language code from ISO 639-1 to ISO 639-2/T""" | |
2957 | return cls._lang_map.get(code[:2]) | |
2958 | ||
2959 | @classmethod | |
2960 | def long2short(cls, code): | |
2961 | """Convert language code from ISO 639-2/T to ISO 639-1""" | |
2962 | for short_name, long_name in cls._lang_map.items(): | |
2963 | if long_name == code: | |
2964 | return short_name | |
2965 | ||
2966 | ||
4eb10f66 YCH |
2967 | class ISO3166Utils(object): |
2968 | # From http://data.okfn.org/data/core/country-list | |
2969 | _country_map = { | |
2970 | 'AF': 'Afghanistan', | |
2971 | 'AX': 'Åland Islands', | |
2972 | 'AL': 'Albania', | |
2973 | 'DZ': 'Algeria', | |
2974 | 'AS': 'American Samoa', | |
2975 | 'AD': 'Andorra', | |
2976 | 'AO': 'Angola', | |
2977 | 'AI': 'Anguilla', | |
2978 | 'AQ': 'Antarctica', | |
2979 | 'AG': 'Antigua and Barbuda', | |
2980 | 'AR': 'Argentina', | |
2981 | 'AM': 'Armenia', | |
2982 | 'AW': 'Aruba', | |
2983 | 'AU': 'Australia', | |
2984 | 'AT': 'Austria', | |
2985 | 'AZ': 'Azerbaijan', | |
2986 | 'BS': 'Bahamas', | |
2987 | 'BH': 'Bahrain', | |
2988 | 'BD': 'Bangladesh', | |
2989 | 'BB': 'Barbados', | |
2990 | 'BY': 'Belarus', | |
2991 | 'BE': 'Belgium', | |
2992 | 'BZ': 'Belize', | |
2993 | 'BJ': 'Benin', | |
2994 | 'BM': 'Bermuda', | |
2995 | 'BT': 'Bhutan', | |
2996 | 'BO': 'Bolivia, Plurinational State of', | |
2997 | 'BQ': 'Bonaire, Sint Eustatius and Saba', | |
2998 | 'BA': 'Bosnia and Herzegovina', | |
2999 | 'BW': 'Botswana', | |
3000 | 'BV': 'Bouvet Island', | |
3001 | 'BR': 'Brazil', | |
3002 | 'IO': 'British Indian Ocean Territory', | |
3003 | 'BN': 'Brunei Darussalam', | |
3004 | 'BG': 'Bulgaria', | |
3005 | 'BF': 'Burkina Faso', | |
3006 | 'BI': 'Burundi', | |
3007 | 'KH': 'Cambodia', | |
3008 | 'CM': 'Cameroon', | |
3009 | 'CA': 'Canada', | |
3010 | 'CV': 'Cape Verde', | |
3011 | 'KY': 'Cayman Islands', | |
3012 | 'CF': 'Central African Republic', | |
3013 | 'TD': 'Chad', | |
3014 | 'CL': 'Chile', | |
3015 | 'CN': 'China', | |
3016 | 'CX': 'Christmas Island', | |
3017 | 'CC': 'Cocos (Keeling) Islands', | |
3018 | 'CO': 'Colombia', | |
3019 | 'KM': 'Comoros', | |
3020 | 'CG': 'Congo', | |
3021 | 'CD': 'Congo, the Democratic Republic of the', | |
3022 | 'CK': 'Cook Islands', | |
3023 | 'CR': 'Costa Rica', | |
3024 | 'CI': 'Côte d\'Ivoire', | |
3025 | 'HR': 'Croatia', | |
3026 | 'CU': 'Cuba', | |
3027 | 'CW': 'Curaçao', | |
3028 | 'CY': 'Cyprus', | |
3029 | 'CZ': 'Czech Republic', | |
3030 | 'DK': 'Denmark', | |
3031 | 'DJ': 'Djibouti', | |
3032 | 'DM': 'Dominica', | |
3033 | 'DO': 'Dominican Republic', | |
3034 | 'EC': 'Ecuador', | |
3035 | 'EG': 'Egypt', | |
3036 | 'SV': 'El Salvador', | |
3037 | 'GQ': 'Equatorial Guinea', | |
3038 | 'ER': 'Eritrea', | |
3039 | 'EE': 'Estonia', | |
3040 | 'ET': 'Ethiopia', | |
3041 | 'FK': 'Falkland Islands (Malvinas)', | |
3042 | 'FO': 'Faroe Islands', | |
3043 | 'FJ': 'Fiji', | |
3044 | 'FI': 'Finland', | |
3045 | 'FR': 'France', | |
3046 | 'GF': 'French Guiana', | |
3047 | 'PF': 'French Polynesia', | |
3048 | 'TF': 'French Southern Territories', | |
3049 | 'GA': 'Gabon', | |
3050 | 'GM': 'Gambia', | |
3051 | 'GE': 'Georgia', | |
3052 | 'DE': 'Germany', | |
3053 | 'GH': 'Ghana', | |
3054 | 'GI': 'Gibraltar', | |
3055 | 'GR': 'Greece', | |
3056 | 'GL': 'Greenland', | |
3057 | 'GD': 'Grenada', | |
3058 | 'GP': 'Guadeloupe', | |
3059 | 'GU': 'Guam', | |
3060 | 'GT': 'Guatemala', | |
3061 | 'GG': 'Guernsey', | |
3062 | 'GN': 'Guinea', | |
3063 | 'GW': 'Guinea-Bissau', | |
3064 | 'GY': 'Guyana', | |
3065 | 'HT': 'Haiti', | |
3066 | 'HM': 'Heard Island and McDonald Islands', | |
3067 | 'VA': 'Holy See (Vatican City State)', | |
3068 | 'HN': 'Honduras', | |
3069 | 'HK': 'Hong Kong', | |
3070 | 'HU': 'Hungary', | |
3071 | 'IS': 'Iceland', | |
3072 | 'IN': 'India', | |
3073 | 'ID': 'Indonesia', | |
3074 | 'IR': 'Iran, Islamic Republic of', | |
3075 | 'IQ': 'Iraq', | |
3076 | 'IE': 'Ireland', | |
3077 | 'IM': 'Isle of Man', | |
3078 | 'IL': 'Israel', | |
3079 | 'IT': 'Italy', | |
3080 | 'JM': 'Jamaica', | |
3081 | 'JP': 'Japan', | |
3082 | 'JE': 'Jersey', | |
3083 | 'JO': 'Jordan', | |
3084 | 'KZ': 'Kazakhstan', | |
3085 | 'KE': 'Kenya', | |
3086 | 'KI': 'Kiribati', | |
3087 | 'KP': 'Korea, Democratic People\'s Republic of', | |
3088 | 'KR': 'Korea, Republic of', | |
3089 | 'KW': 'Kuwait', | |
3090 | 'KG': 'Kyrgyzstan', | |
3091 | 'LA': 'Lao People\'s Democratic Republic', | |
3092 | 'LV': 'Latvia', | |
3093 | 'LB': 'Lebanon', | |
3094 | 'LS': 'Lesotho', | |
3095 | 'LR': 'Liberia', | |
3096 | 'LY': 'Libya', | |
3097 | 'LI': 'Liechtenstein', | |
3098 | 'LT': 'Lithuania', | |
3099 | 'LU': 'Luxembourg', | |
3100 | 'MO': 'Macao', | |
3101 | 'MK': 'Macedonia, the Former Yugoslav Republic of', | |
3102 | 'MG': 'Madagascar', | |
3103 | 'MW': 'Malawi', | |
3104 | 'MY': 'Malaysia', | |
3105 | 'MV': 'Maldives', | |
3106 | 'ML': 'Mali', | |
3107 | 'MT': 'Malta', | |
3108 | 'MH': 'Marshall Islands', | |
3109 | 'MQ': 'Martinique', | |
3110 | 'MR': 'Mauritania', | |
3111 | 'MU': 'Mauritius', | |
3112 | 'YT': 'Mayotte', | |
3113 | 'MX': 'Mexico', | |
3114 | 'FM': 'Micronesia, Federated States of', | |
3115 | 'MD': 'Moldova, Republic of', | |
3116 | 'MC': 'Monaco', | |
3117 | 'MN': 'Mongolia', | |
3118 | 'ME': 'Montenegro', | |
3119 | 'MS': 'Montserrat', | |
3120 | 'MA': 'Morocco', | |
3121 | 'MZ': 'Mozambique', | |
3122 | 'MM': 'Myanmar', | |
3123 | 'NA': 'Namibia', | |
3124 | 'NR': 'Nauru', | |
3125 | 'NP': 'Nepal', | |
3126 | 'NL': 'Netherlands', | |
3127 | 'NC': 'New Caledonia', | |
3128 | 'NZ': 'New Zealand', | |
3129 | 'NI': 'Nicaragua', | |
3130 | 'NE': 'Niger', | |
3131 | 'NG': 'Nigeria', | |
3132 | 'NU': 'Niue', | |
3133 | 'NF': 'Norfolk Island', | |
3134 | 'MP': 'Northern Mariana Islands', | |
3135 | 'NO': 'Norway', | |
3136 | 'OM': 'Oman', | |
3137 | 'PK': 'Pakistan', | |
3138 | 'PW': 'Palau', | |
3139 | 'PS': 'Palestine, State of', | |
3140 | 'PA': 'Panama', | |
3141 | 'PG': 'Papua New Guinea', | |
3142 | 'PY': 'Paraguay', | |
3143 | 'PE': 'Peru', | |
3144 | 'PH': 'Philippines', | |
3145 | 'PN': 'Pitcairn', | |
3146 | 'PL': 'Poland', | |
3147 | 'PT': 'Portugal', | |
3148 | 'PR': 'Puerto Rico', | |
3149 | 'QA': 'Qatar', | |
3150 | 'RE': 'Réunion', | |
3151 | 'RO': 'Romania', | |
3152 | 'RU': 'Russian Federation', | |
3153 | 'RW': 'Rwanda', | |
3154 | 'BL': 'Saint Barthélemy', | |
3155 | 'SH': 'Saint Helena, Ascension and Tristan da Cunha', | |
3156 | 'KN': 'Saint Kitts and Nevis', | |
3157 | 'LC': 'Saint Lucia', | |
3158 | 'MF': 'Saint Martin (French part)', | |
3159 | 'PM': 'Saint Pierre and Miquelon', | |
3160 | 'VC': 'Saint Vincent and the Grenadines', | |
3161 | 'WS': 'Samoa', | |
3162 | 'SM': 'San Marino', | |
3163 | 'ST': 'Sao Tome and Principe', | |
3164 | 'SA': 'Saudi Arabia', | |
3165 | 'SN': 'Senegal', | |
3166 | 'RS': 'Serbia', | |
3167 | 'SC': 'Seychelles', | |
3168 | 'SL': 'Sierra Leone', | |
3169 | 'SG': 'Singapore', | |
3170 | 'SX': 'Sint Maarten (Dutch part)', | |
3171 | 'SK': 'Slovakia', | |
3172 | 'SI': 'Slovenia', | |
3173 | 'SB': 'Solomon Islands', | |
3174 | 'SO': 'Somalia', | |
3175 | 'ZA': 'South Africa', | |
3176 | 'GS': 'South Georgia and the South Sandwich Islands', | |
3177 | 'SS': 'South Sudan', | |
3178 | 'ES': 'Spain', | |
3179 | 'LK': 'Sri Lanka', | |
3180 | 'SD': 'Sudan', | |
3181 | 'SR': 'Suriname', | |
3182 | 'SJ': 'Svalbard and Jan Mayen', | |
3183 | 'SZ': 'Swaziland', | |
3184 | 'SE': 'Sweden', | |
3185 | 'CH': 'Switzerland', | |
3186 | 'SY': 'Syrian Arab Republic', | |
3187 | 'TW': 'Taiwan, Province of China', | |
3188 | 'TJ': 'Tajikistan', | |
3189 | 'TZ': 'Tanzania, United Republic of', | |
3190 | 'TH': 'Thailand', | |
3191 | 'TL': 'Timor-Leste', | |
3192 | 'TG': 'Togo', | |
3193 | 'TK': 'Tokelau', | |
3194 | 'TO': 'Tonga', | |
3195 | 'TT': 'Trinidad and Tobago', | |
3196 | 'TN': 'Tunisia', | |
3197 | 'TR': 'Turkey', | |
3198 | 'TM': 'Turkmenistan', | |
3199 | 'TC': 'Turks and Caicos Islands', | |
3200 | 'TV': 'Tuvalu', | |
3201 | 'UG': 'Uganda', | |
3202 | 'UA': 'Ukraine', | |
3203 | 'AE': 'United Arab Emirates', | |
3204 | 'GB': 'United Kingdom', | |
3205 | 'US': 'United States', | |
3206 | 'UM': 'United States Minor Outlying Islands', | |
3207 | 'UY': 'Uruguay', | |
3208 | 'UZ': 'Uzbekistan', | |
3209 | 'VU': 'Vanuatu', | |
3210 | 'VE': 'Venezuela, Bolivarian Republic of', | |
3211 | 'VN': 'Viet Nam', | |
3212 | 'VG': 'Virgin Islands, British', | |
3213 | 'VI': 'Virgin Islands, U.S.', | |
3214 | 'WF': 'Wallis and Futuna', | |
3215 | 'EH': 'Western Sahara', | |
3216 | 'YE': 'Yemen', | |
3217 | 'ZM': 'Zambia', | |
3218 | 'ZW': 'Zimbabwe', | |
3219 | } | |
3220 | ||
3221 | @classmethod | |
3222 | def short2full(cls, code): | |
3223 | """Convert an ISO 3166-2 country code to the corresponding full name""" | |
3224 | return cls._country_map.get(code.upper()) | |
3225 | ||
3226 | ||
773f291d S |
3227 | class GeoUtils(object): |
3228 | # Major IPv4 address blocks per country | |
3229 | _country_ip_map = { | |
3230 | 'AD': '85.94.160.0/19', | |
3231 | 'AE': '94.200.0.0/13', | |
3232 | 'AF': '149.54.0.0/17', | |
3233 | 'AG': '209.59.64.0/18', | |
3234 | 'AI': '204.14.248.0/21', | |
3235 | 'AL': '46.99.0.0/16', | |
3236 | 'AM': '46.70.0.0/15', | |
3237 | 'AO': '105.168.0.0/13', | |
3238 | 'AP': '159.117.192.0/21', | |
3239 | 'AR': '181.0.0.0/12', | |
3240 | 'AS': '202.70.112.0/20', | |
3241 | 'AT': '84.112.0.0/13', | |
3242 | 'AU': '1.128.0.0/11', | |
3243 | 'AW': '181.41.0.0/18', | |
3244 | 'AZ': '5.191.0.0/16', | |
3245 | 'BA': '31.176.128.0/17', | |
3246 | 'BB': '65.48.128.0/17', | |
3247 | 'BD': '114.130.0.0/16', | |
3248 | 'BE': '57.0.0.0/8', | |
3249 | 'BF': '129.45.128.0/17', | |
3250 | 'BG': '95.42.0.0/15', | |
3251 | 'BH': '37.131.0.0/17', | |
3252 | 'BI': '154.117.192.0/18', | |
3253 | 'BJ': '137.255.0.0/16', | |
3254 | 'BL': '192.131.134.0/24', | |
3255 | 'BM': '196.12.64.0/18', | |
3256 | 'BN': '156.31.0.0/16', | |
3257 | 'BO': '161.56.0.0/16', | |
3258 | 'BQ': '161.0.80.0/20', | |
3259 | 'BR': '152.240.0.0/12', | |
3260 | 'BS': '24.51.64.0/18', | |
3261 | 'BT': '119.2.96.0/19', | |
3262 | 'BW': '168.167.0.0/16', | |
3263 | 'BY': '178.120.0.0/13', | |
3264 | 'BZ': '179.42.192.0/18', | |
3265 | 'CA': '99.224.0.0/11', | |
3266 | 'CD': '41.243.0.0/16', | |
3267 | 'CF': '196.32.200.0/21', | |
3268 | 'CG': '197.214.128.0/17', | |
3269 | 'CH': '85.0.0.0/13', | |
3270 | 'CI': '154.232.0.0/14', | |
3271 | 'CK': '202.65.32.0/19', | |
3272 | 'CL': '152.172.0.0/14', | |
3273 | 'CM': '165.210.0.0/15', | |
3274 | 'CN': '36.128.0.0/10', | |
3275 | 'CO': '181.240.0.0/12', | |
3276 | 'CR': '201.192.0.0/12', | |
3277 | 'CU': '152.206.0.0/15', | |
3278 | 'CV': '165.90.96.0/19', | |
3279 | 'CW': '190.88.128.0/17', | |
3280 | 'CY': '46.198.0.0/15', | |
3281 | 'CZ': '88.100.0.0/14', | |
3282 | 'DE': '53.0.0.0/8', | |
3283 | 'DJ': '197.241.0.0/17', | |
3284 | 'DK': '87.48.0.0/12', | |
3285 | 'DM': '192.243.48.0/20', | |
3286 | 'DO': '152.166.0.0/15', | |
3287 | 'DZ': '41.96.0.0/12', | |
3288 | 'EC': '186.68.0.0/15', | |
3289 | 'EE': '90.190.0.0/15', | |
3290 | 'EG': '156.160.0.0/11', | |
3291 | 'ER': '196.200.96.0/20', | |
3292 | 'ES': '88.0.0.0/11', | |
3293 | 'ET': '196.188.0.0/14', | |
3294 | 'EU': '2.16.0.0/13', | |
3295 | 'FI': '91.152.0.0/13', | |
3296 | 'FJ': '144.120.0.0/16', | |
3297 | 'FM': '119.252.112.0/20', | |
3298 | 'FO': '88.85.32.0/19', | |
3299 | 'FR': '90.0.0.0/9', | |
3300 | 'GA': '41.158.0.0/15', | |
3301 | 'GB': '25.0.0.0/8', | |
3302 | 'GD': '74.122.88.0/21', | |
3303 | 'GE': '31.146.0.0/16', | |
3304 | 'GF': '161.22.64.0/18', | |
3305 | 'GG': '62.68.160.0/19', | |
3306 | 'GH': '45.208.0.0/14', | |
3307 | 'GI': '85.115.128.0/19', | |
3308 | 'GL': '88.83.0.0/19', | |
3309 | 'GM': '160.182.0.0/15', | |
3310 | 'GN': '197.149.192.0/18', | |
3311 | 'GP': '104.250.0.0/19', | |
3312 | 'GQ': '105.235.224.0/20', | |
3313 | 'GR': '94.64.0.0/13', | |
3314 | 'GT': '168.234.0.0/16', | |
3315 | 'GU': '168.123.0.0/16', | |
3316 | 'GW': '197.214.80.0/20', | |
3317 | 'GY': '181.41.64.0/18', | |
3318 | 'HK': '113.252.0.0/14', | |
3319 | 'HN': '181.210.0.0/16', | |
3320 | 'HR': '93.136.0.0/13', | |
3321 | 'HT': '148.102.128.0/17', | |
3322 | 'HU': '84.0.0.0/14', | |
3323 | 'ID': '39.192.0.0/10', | |
3324 | 'IE': '87.32.0.0/12', | |
3325 | 'IL': '79.176.0.0/13', | |
3326 | 'IM': '5.62.80.0/20', | |
3327 | 'IN': '117.192.0.0/10', | |
3328 | 'IO': '203.83.48.0/21', | |
3329 | 'IQ': '37.236.0.0/14', | |
3330 | 'IR': '2.176.0.0/12', | |
3331 | 'IS': '82.221.0.0/16', | |
3332 | 'IT': '79.0.0.0/10', | |
3333 | 'JE': '87.244.64.0/18', | |
3334 | 'JM': '72.27.0.0/17', | |
3335 | 'JO': '176.29.0.0/16', | |
3336 | 'JP': '126.0.0.0/8', | |
3337 | 'KE': '105.48.0.0/12', | |
3338 | 'KG': '158.181.128.0/17', | |
3339 | 'KH': '36.37.128.0/17', | |
3340 | 'KI': '103.25.140.0/22', | |
3341 | 'KM': '197.255.224.0/20', | |
3342 | 'KN': '198.32.32.0/19', | |
3343 | 'KP': '175.45.176.0/22', | |
3344 | 'KR': '175.192.0.0/10', | |
3345 | 'KW': '37.36.0.0/14', | |
3346 | 'KY': '64.96.0.0/15', | |
3347 | 'KZ': '2.72.0.0/13', | |
3348 | 'LA': '115.84.64.0/18', | |
3349 | 'LB': '178.135.0.0/16', | |
3350 | 'LC': '192.147.231.0/24', | |
3351 | 'LI': '82.117.0.0/19', | |
3352 | 'LK': '112.134.0.0/15', | |
3353 | 'LR': '41.86.0.0/19', | |
3354 | 'LS': '129.232.0.0/17', | |
3355 | 'LT': '78.56.0.0/13', | |
3356 | 'LU': '188.42.0.0/16', | |
3357 | 'LV': '46.109.0.0/16', | |
3358 | 'LY': '41.252.0.0/14', | |
3359 | 'MA': '105.128.0.0/11', | |
3360 | 'MC': '88.209.64.0/18', | |
3361 | 'MD': '37.246.0.0/16', | |
3362 | 'ME': '178.175.0.0/17', | |
3363 | 'MF': '74.112.232.0/21', | |
3364 | 'MG': '154.126.0.0/17', | |
3365 | 'MH': '117.103.88.0/21', | |
3366 | 'MK': '77.28.0.0/15', | |
3367 | 'ML': '154.118.128.0/18', | |
3368 | 'MM': '37.111.0.0/17', | |
3369 | 'MN': '49.0.128.0/17', | |
3370 | 'MO': '60.246.0.0/16', | |
3371 | 'MP': '202.88.64.0/20', | |
3372 | 'MQ': '109.203.224.0/19', | |
3373 | 'MR': '41.188.64.0/18', | |
3374 | 'MS': '208.90.112.0/22', | |
3375 | 'MT': '46.11.0.0/16', | |
3376 | 'MU': '105.16.0.0/12', | |
3377 | 'MV': '27.114.128.0/18', | |
3378 | 'MW': '105.234.0.0/16', | |
3379 | 'MX': '187.192.0.0/11', | |
3380 | 'MY': '175.136.0.0/13', | |
3381 | 'MZ': '197.218.0.0/15', | |
3382 | 'NA': '41.182.0.0/16', | |
3383 | 'NC': '101.101.0.0/18', | |
3384 | 'NE': '197.214.0.0/18', | |
3385 | 'NF': '203.17.240.0/22', | |
3386 | 'NG': '105.112.0.0/12', | |
3387 | 'NI': '186.76.0.0/15', | |
3388 | 'NL': '145.96.0.0/11', | |
3389 | 'NO': '84.208.0.0/13', | |
3390 | 'NP': '36.252.0.0/15', | |
3391 | 'NR': '203.98.224.0/19', | |
3392 | 'NU': '49.156.48.0/22', | |
3393 | 'NZ': '49.224.0.0/14', | |
3394 | 'OM': '5.36.0.0/15', | |
3395 | 'PA': '186.72.0.0/15', | |
3396 | 'PE': '186.160.0.0/14', | |
3397 | 'PF': '123.50.64.0/18', | |
3398 | 'PG': '124.240.192.0/19', | |
3399 | 'PH': '49.144.0.0/13', | |
3400 | 'PK': '39.32.0.0/11', | |
3401 | 'PL': '83.0.0.0/11', | |
3402 | 'PM': '70.36.0.0/20', | |
3403 | 'PR': '66.50.0.0/16', | |
3404 | 'PS': '188.161.0.0/16', | |
3405 | 'PT': '85.240.0.0/13', | |
3406 | 'PW': '202.124.224.0/20', | |
3407 | 'PY': '181.120.0.0/14', | |
3408 | 'QA': '37.210.0.0/15', | |
3409 | 'RE': '139.26.0.0/16', | |
3410 | 'RO': '79.112.0.0/13', | |
3411 | 'RS': '178.220.0.0/14', | |
3412 | 'RU': '5.136.0.0/13', | |
3413 | 'RW': '105.178.0.0/15', | |
3414 | 'SA': '188.48.0.0/13', | |
3415 | 'SB': '202.1.160.0/19', | |
3416 | 'SC': '154.192.0.0/11', | |
3417 | 'SD': '154.96.0.0/13', | |
3418 | 'SE': '78.64.0.0/12', | |
3419 | 'SG': '152.56.0.0/14', | |
3420 | 'SI': '188.196.0.0/14', | |
3421 | 'SK': '78.98.0.0/15', | |
3422 | 'SL': '197.215.0.0/17', | |
3423 | 'SM': '89.186.32.0/19', | |
3424 | 'SN': '41.82.0.0/15', | |
3425 | 'SO': '197.220.64.0/19', | |
3426 | 'SR': '186.179.128.0/17', | |
3427 | 'SS': '105.235.208.0/21', | |
3428 | 'ST': '197.159.160.0/19', | |
3429 | 'SV': '168.243.0.0/16', | |
3430 | 'SX': '190.102.0.0/20', | |
3431 | 'SY': '5.0.0.0/16', | |
3432 | 'SZ': '41.84.224.0/19', | |
3433 | 'TC': '65.255.48.0/20', | |
3434 | 'TD': '154.68.128.0/19', | |
3435 | 'TG': '196.168.0.0/14', | |
3436 | 'TH': '171.96.0.0/13', | |
3437 | 'TJ': '85.9.128.0/18', | |
3438 | 'TK': '27.96.24.0/21', | |
3439 | 'TL': '180.189.160.0/20', | |
3440 | 'TM': '95.85.96.0/19', | |
3441 | 'TN': '197.0.0.0/11', | |
3442 | 'TO': '175.176.144.0/21', | |
3443 | 'TR': '78.160.0.0/11', | |
3444 | 'TT': '186.44.0.0/15', | |
3445 | 'TV': '202.2.96.0/19', | |
3446 | 'TW': '120.96.0.0/11', | |
3447 | 'TZ': '156.156.0.0/14', | |
3448 | 'UA': '93.72.0.0/13', | |
3449 | 'UG': '154.224.0.0/13', | |
3450 | 'US': '3.0.0.0/8', | |
3451 | 'UY': '167.56.0.0/13', | |
3452 | 'UZ': '82.215.64.0/18', | |
3453 | 'VA': '212.77.0.0/19', | |
3454 | 'VC': '24.92.144.0/20', | |
3455 | 'VE': '186.88.0.0/13', | |
3456 | 'VG': '172.103.64.0/18', | |
3457 | 'VI': '146.226.0.0/16', | |
3458 | 'VN': '14.160.0.0/11', | |
3459 | 'VU': '202.80.32.0/20', | |
3460 | 'WF': '117.20.32.0/21', | |
3461 | 'WS': '202.4.32.0/19', | |
3462 | 'YE': '134.35.0.0/16', | |
3463 | 'YT': '41.242.116.0/22', | |
3464 | 'ZA': '41.0.0.0/11', | |
3465 | 'ZM': '165.56.0.0/13', | |
3466 | 'ZW': '41.85.192.0/19', | |
3467 | } | |
3468 | ||
3469 | @classmethod | |
3470 | def random_ipv4(cls, code): | |
3471 | block = cls._country_ip_map.get(code.upper()) | |
3472 | if not block: | |
3473 | return None | |
3474 | addr, preflen = block.split('/') | |
3475 | addr_min = compat_struct_unpack('!L', socket.inet_aton(addr))[0] | |
3476 | addr_max = addr_min | (0xffffffff >> int(preflen)) | |
18a0defa | 3477 | return compat_str(socket.inet_ntoa( |
4248dad9 | 3478 | compat_struct_pack('!L', random.randint(addr_min, addr_max)))) |
773f291d S |
3479 | |
3480 | ||
91410c9b | 3481 | class PerRequestProxyHandler(compat_urllib_request.ProxyHandler): |
2461f79d PH |
3482 | def __init__(self, proxies=None): |
3483 | # Set default handlers | |
3484 | for type in ('http', 'https'): | |
3485 | setattr(self, '%s_open' % type, | |
3486 | lambda r, proxy='__noproxy__', type=type, meth=self.proxy_open: | |
3487 | meth(r, proxy, type)) | |
3488 | return compat_urllib_request.ProxyHandler.__init__(self, proxies) | |
3489 | ||
91410c9b | 3490 | def proxy_open(self, req, proxy, type): |
2461f79d | 3491 | req_proxy = req.headers.get('Ytdl-request-proxy') |
91410c9b PH |
3492 | if req_proxy is not None: |
3493 | proxy = req_proxy | |
2461f79d PH |
3494 | del req.headers['Ytdl-request-proxy'] |
3495 | ||
3496 | if proxy == '__noproxy__': | |
3497 | return None # No Proxy | |
51fb4995 | 3498 | if compat_urlparse.urlparse(proxy).scheme.lower() in ('socks', 'socks4', 'socks4a', 'socks5'): |
71aff188 YCH |
3499 | req.add_header('Ytdl-socks-proxy', proxy) |
3500 | # youtube-dl's http/https handlers do wrapping the socket with socks | |
3501 | return None | |
91410c9b PH |
3502 | return compat_urllib_request.ProxyHandler.proxy_open( |
3503 | self, req, proxy, type) | |
5bc880b9 YCH |
3504 | |
3505 | ||
0a5445dd YCH |
3506 | # Both long_to_bytes and bytes_to_long are adapted from PyCrypto, which is |
3507 | # released into Public Domain | |
3508 | # https://github.com/dlitz/pycrypto/blob/master/lib/Crypto/Util/number.py#L387 | |
3509 | ||
3510 | def long_to_bytes(n, blocksize=0): | |
3511 | """long_to_bytes(n:long, blocksize:int) : string | |
3512 | Convert a long integer to a byte string. | |
3513 | ||
3514 | If optional blocksize is given and greater than zero, pad the front of the | |
3515 | byte string with binary zeros so that the length is a multiple of | |
3516 | blocksize. | |
3517 | """ | |
3518 | # after much testing, this algorithm was deemed to be the fastest | |
3519 | s = b'' | |
3520 | n = int(n) | |
3521 | while n > 0: | |
3522 | s = compat_struct_pack('>I', n & 0xffffffff) + s | |
3523 | n = n >> 32 | |
3524 | # strip off leading zeros | |
3525 | for i in range(len(s)): | |
3526 | if s[i] != b'\000'[0]: | |
3527 | break | |
3528 | else: | |
3529 | # only happens when n == 0 | |
3530 | s = b'\000' | |
3531 | i = 0 | |
3532 | s = s[i:] | |
3533 | # add back some pad bytes. this could be done more efficiently w.r.t. the | |
3534 | # de-padding being done above, but sigh... | |
3535 | if blocksize > 0 and len(s) % blocksize: | |
3536 | s = (blocksize - len(s) % blocksize) * b'\000' + s | |
3537 | return s | |
3538 | ||
3539 | ||
3540 | def bytes_to_long(s): | |
3541 | """bytes_to_long(string) : long | |
3542 | Convert a byte string to a long integer. | |
3543 | ||
3544 | This is (essentially) the inverse of long_to_bytes(). | |
3545 | """ | |
3546 | acc = 0 | |
3547 | length = len(s) | |
3548 | if length % 4: | |
3549 | extra = (4 - length % 4) | |
3550 | s = b'\000' * extra + s | |
3551 | length = length + extra | |
3552 | for i in range(0, length, 4): | |
3553 | acc = (acc << 32) + compat_struct_unpack('>I', s[i:i + 4])[0] | |
3554 | return acc | |
3555 | ||
3556 | ||
5bc880b9 YCH |
3557 | def ohdave_rsa_encrypt(data, exponent, modulus): |
3558 | ''' | |
3559 | Implement OHDave's RSA algorithm. See http://www.ohdave.com/rsa/ | |
3560 | ||
3561 | Input: | |
3562 | data: data to encrypt, bytes-like object | |
3563 | exponent, modulus: parameter e and N of RSA algorithm, both integer | |
3564 | Output: hex string of encrypted data | |
3565 | ||
3566 | Limitation: supports one block encryption only | |
3567 | ''' | |
3568 | ||
3569 | payload = int(binascii.hexlify(data[::-1]), 16) | |
3570 | encrypted = pow(payload, exponent, modulus) | |
3571 | return '%x' % encrypted | |
81bdc8fd YCH |
3572 | |
3573 | ||
f48409c7 YCH |
3574 | def pkcs1pad(data, length): |
3575 | """ | |
3576 | Padding input data with PKCS#1 scheme | |
3577 | ||
3578 | @param {int[]} data input data | |
3579 | @param {int} length target length | |
3580 | @returns {int[]} padded data | |
3581 | """ | |
3582 | if len(data) > length - 11: | |
3583 | raise ValueError('Input data too long for PKCS#1 padding') | |
3584 | ||
3585 | pseudo_random = [random.randint(0, 254) for _ in range(length - len(data) - 3)] | |
3586 | return [0, 2] + pseudo_random + [0] + data | |
3587 | ||
3588 | ||
5eb6bdce | 3589 | def encode_base_n(num, n, table=None): |
59f898b7 | 3590 | FULL_TABLE = '0123456789abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ' |
59f898b7 YCH |
3591 | if not table: |
3592 | table = FULL_TABLE[:n] | |
3593 | ||
5eb6bdce YCH |
3594 | if n > len(table): |
3595 | raise ValueError('base %d exceeds table length %d' % (n, len(table))) | |
3596 | ||
3597 | if num == 0: | |
3598 | return table[0] | |
3599 | ||
81bdc8fd YCH |
3600 | ret = '' |
3601 | while num: | |
3602 | ret = table[num % n] + ret | |
3603 | num = num // n | |
3604 | return ret | |
f52354a8 YCH |
3605 | |
3606 | ||
3607 | def decode_packed_codes(code): | |
06b3fe29 | 3608 | mobj = re.search(PACKED_CODES_RE, code) |
f52354a8 YCH |
3609 | obfucasted_code, base, count, symbols = mobj.groups() |
3610 | base = int(base) | |
3611 | count = int(count) | |
3612 | symbols = symbols.split('|') | |
3613 | symbol_table = {} | |
3614 | ||
3615 | while count: | |
3616 | count -= 1 | |
5eb6bdce | 3617 | base_n_count = encode_base_n(count, base) |
f52354a8 YCH |
3618 | symbol_table[base_n_count] = symbols[count] or base_n_count |
3619 | ||
3620 | return re.sub( | |
3621 | r'\b(\w+)\b', lambda mobj: symbol_table[mobj.group(0)], | |
3622 | obfucasted_code) | |
e154c651 | 3623 | |
3624 | ||
3625 | def parse_m3u8_attributes(attrib): | |
3626 | info = {} | |
3627 | for (key, val) in re.findall(r'(?P<key>[A-Z0-9-]+)=(?P<val>"[^"]+"|[^",]+)(?:,|$)', attrib): | |
3628 | if val.startswith('"'): | |
3629 | val = val[1:-1] | |
3630 | info[key] = val | |
3631 | return info | |
1143535d YCH |
3632 | |
3633 | ||
3634 | def urshift(val, n): | |
3635 | return val >> n if val >= 0 else (val + 0x100000000) >> n | |
d3f8e038 YCH |
3636 | |
3637 | ||
3638 | # Based on png2str() written by @gdkchan and improved by @yokrysty | |
3639 | # Originally posted at https://github.com/rg3/youtube-dl/issues/9706 | |
3640 | def decode_png(png_data): | |
3641 | # Reference: https://www.w3.org/TR/PNG/ | |
3642 | header = png_data[8:] | |
3643 | ||
3644 | if png_data[:8] != b'\x89PNG\x0d\x0a\x1a\x0a' or header[4:8] != b'IHDR': | |
3645 | raise IOError('Not a valid PNG file.') | |
3646 | ||
3647 | int_map = {1: '>B', 2: '>H', 4: '>I'} | |
3648 | unpack_integer = lambda x: compat_struct_unpack(int_map[len(x)], x)[0] | |
3649 | ||
3650 | chunks = [] | |
3651 | ||
3652 | while header: | |
3653 | length = unpack_integer(header[:4]) | |
3654 | header = header[4:] | |
3655 | ||
3656 | chunk_type = header[:4] | |
3657 | header = header[4:] | |
3658 | ||
3659 | chunk_data = header[:length] | |
3660 | header = header[length:] | |
3661 | ||
3662 | header = header[4:] # Skip CRC | |
3663 | ||
3664 | chunks.append({ | |
3665 | 'type': chunk_type, | |
3666 | 'length': length, | |
3667 | 'data': chunk_data | |
3668 | }) | |
3669 | ||
3670 | ihdr = chunks[0]['data'] | |
3671 | ||
3672 | width = unpack_integer(ihdr[:4]) | |
3673 | height = unpack_integer(ihdr[4:8]) | |
3674 | ||
3675 | idat = b'' | |
3676 | ||
3677 | for chunk in chunks: | |
3678 | if chunk['type'] == b'IDAT': | |
3679 | idat += chunk['data'] | |
3680 | ||
3681 | if not idat: | |
3682 | raise IOError('Unable to read PNG data.') | |
3683 | ||
3684 | decompressed_data = bytearray(zlib.decompress(idat)) | |
3685 | ||
3686 | stride = width * 3 | |
3687 | pixels = [] | |
3688 | ||
3689 | def _get_pixel(idx): | |
3690 | x = idx % stride | |
3691 | y = idx // stride | |
3692 | return pixels[y][x] | |
3693 | ||
3694 | for y in range(height): | |
3695 | basePos = y * (1 + stride) | |
3696 | filter_type = decompressed_data[basePos] | |
3697 | ||
3698 | current_row = [] | |
3699 | ||
3700 | pixels.append(current_row) | |
3701 | ||
3702 | for x in range(stride): | |
3703 | color = decompressed_data[1 + basePos + x] | |
3704 | basex = y * stride + x | |
3705 | left = 0 | |
3706 | up = 0 | |
3707 | ||
3708 | if x > 2: | |
3709 | left = _get_pixel(basex - 3) | |
3710 | if y > 0: | |
3711 | up = _get_pixel(basex - stride) | |
3712 | ||
3713 | if filter_type == 1: # Sub | |
3714 | color = (color + left) & 0xff | |
3715 | elif filter_type == 2: # Up | |
3716 | color = (color + up) & 0xff | |
3717 | elif filter_type == 3: # Average | |
3718 | color = (color + ((left + up) >> 1)) & 0xff | |
3719 | elif filter_type == 4: # Paeth | |
3720 | a = left | |
3721 | b = up | |
3722 | c = 0 | |
3723 | ||
3724 | if x > 2 and y > 0: | |
3725 | c = _get_pixel(basex - stride - 3) | |
3726 | ||
3727 | p = a + b - c | |
3728 | ||
3729 | pa = abs(p - a) | |
3730 | pb = abs(p - b) | |
3731 | pc = abs(p - c) | |
3732 | ||
3733 | if pa <= pb and pa <= pc: | |
3734 | color = (color + a) & 0xff | |
3735 | elif pb <= pc: | |
3736 | color = (color + b) & 0xff | |
3737 | else: | |
3738 | color = (color + c) & 0xff | |
3739 | ||
3740 | current_row.append(color) | |
3741 | ||
3742 | return width, height, pixels | |
efa97bdc YCH |
3743 | |
3744 | ||
3745 | def write_xattr(path, key, value): | |
3746 | # This mess below finds the best xattr tool for the job | |
3747 | try: | |
3748 | # try the pyxattr module... | |
3749 | import xattr | |
3750 | ||
53a7e3d2 YCH |
3751 | if hasattr(xattr, 'set'): # pyxattr |
3752 | # Unicode arguments are not supported in python-pyxattr until | |
3753 | # version 0.5.0 | |
3754 | # See https://github.com/rg3/youtube-dl/issues/5498 | |
3755 | pyxattr_required_version = '0.5.0' | |
3756 | if version_tuple(xattr.__version__) < version_tuple(pyxattr_required_version): | |
3757 | # TODO: fallback to CLI tools | |
3758 | raise XAttrUnavailableError( | |
3759 | 'python-pyxattr is detected but is too old. ' | |
3760 | 'youtube-dl requires %s or above while your version is %s. ' | |
3761 | 'Falling back to other xattr implementations' % ( | |
3762 | pyxattr_required_version, xattr.__version__)) | |
3763 | ||
3764 | setxattr = xattr.set | |
3765 | else: # xattr | |
3766 | setxattr = xattr.setxattr | |
efa97bdc YCH |
3767 | |
3768 | try: | |
53a7e3d2 | 3769 | setxattr(path, key, value) |
efa97bdc YCH |
3770 | except EnvironmentError as e: |
3771 | raise XAttrMetadataError(e.errno, e.strerror) | |
3772 | ||
3773 | except ImportError: | |
3774 | if compat_os_name == 'nt': | |
3775 | # Write xattrs to NTFS Alternate Data Streams: | |
3776 | # http://en.wikipedia.org/wiki/NTFS#Alternate_data_streams_.28ADS.29 | |
3777 | assert ':' not in key | |
3778 | assert os.path.exists(path) | |
3779 | ||
3780 | ads_fn = path + ':' + key | |
3781 | try: | |
3782 | with open(ads_fn, 'wb') as f: | |
3783 | f.write(value) | |
3784 | except EnvironmentError as e: | |
3785 | raise XAttrMetadataError(e.errno, e.strerror) | |
3786 | else: | |
3787 | user_has_setfattr = check_executable('setfattr', ['--version']) | |
3788 | user_has_xattr = check_executable('xattr', ['-h']) | |
3789 | ||
3790 | if user_has_setfattr or user_has_xattr: | |
3791 | ||
3792 | value = value.decode('utf-8') | |
3793 | if user_has_setfattr: | |
3794 | executable = 'setfattr' | |
3795 | opts = ['-n', key, '-v', value] | |
3796 | elif user_has_xattr: | |
3797 | executable = 'xattr' | |
3798 | opts = ['-w', key, value] | |
3799 | ||
3800 | cmd = ([encodeFilename(executable, True)] + | |
3801 | [encodeArgument(o) for o in opts] + | |
3802 | [encodeFilename(path, True)]) | |
3803 | ||
3804 | try: | |
3805 | p = subprocess.Popen( | |
3806 | cmd, stdout=subprocess.PIPE, stderr=subprocess.PIPE, stdin=subprocess.PIPE) | |
3807 | except EnvironmentError as e: | |
3808 | raise XAttrMetadataError(e.errno, e.strerror) | |
3809 | stdout, stderr = p.communicate() | |
3810 | stderr = stderr.decode('utf-8', 'replace') | |
3811 | if p.returncode != 0: | |
3812 | raise XAttrMetadataError(p.returncode, stderr) | |
3813 | ||
3814 | else: | |
3815 | # On Unix, and can't find pyxattr, setfattr, or xattr. | |
3816 | if sys.platform.startswith('linux'): | |
3817 | raise XAttrUnavailableError( | |
3818 | "Couldn't find a tool to set the xattrs. " | |
3819 | "Install either the python 'pyxattr' or 'xattr' " | |
3820 | "modules, or the GNU 'attr' package " | |
3821 | "(which contains the 'setfattr' tool).") | |
3822 | else: | |
3823 | raise XAttrUnavailableError( | |
3824 | "Couldn't find a tool to set the xattrs. " | |
3825 | "Install either the python 'xattr' module, " | |
3826 | "or the 'xattr' binary.") | |
0c265486 YCH |
3827 | |
3828 | ||
40e41780 TF |
3829 | def cookie_to_dict(cookie): |
3830 | cookie_dict = { | |
3831 | 'name': cookie.name, | |
3832 | 'value': cookie.value, | |
4ed2d7b7 | 3833 | } |
40e41780 TF |
3834 | if cookie.port_specified: |
3835 | cookie_dict['port'] = cookie.port | |
3836 | if cookie.domain_specified: | |
3837 | cookie_dict['domain'] = cookie.domain | |
3838 | if cookie.path_specified: | |
3839 | cookie_dict['path'] = cookie.path | |
4ed2d7b7 | 3840 | if cookie.expires is not None: |
40e41780 | 3841 | cookie_dict['expires'] = cookie.expires |
4ed2d7b7 | 3842 | if cookie.secure is not None: |
40e41780 | 3843 | cookie_dict['secure'] = cookie.secure |
4ed2d7b7 | 3844 | if cookie.discard is not None: |
40e41780 TF |
3845 | cookie_dict['discard'] = cookie.discard |
3846 | try: | |
3847 | if (cookie.has_nonstandard_attr('httpOnly') or | |
4ed2d7b7 YCH |
3848 | cookie.has_nonstandard_attr('httponly') or |
3849 | cookie.has_nonstandard_attr('HttpOnly')): | |
40e41780 TF |
3850 | cookie_dict['httponly'] = True |
3851 | except TypeError: | |
3852 | pass | |
3853 | return cookie_dict | |
3854 | ||
3855 | ||
3856 | def cookie_jar_to_list(cookie_jar): | |
3857 | return [cookie_to_dict(cookie) for cookie in cookie_jar] | |
3858 | ||
3859 | ||
da57ebaf TF |
3860 | class PhantomJSwrapper(object): |
3861 | """PhantomJS wrapper class""" | |
3862 | ||
3863 | _TEMPLATE = r''' | |
3864 | phantom.onError = function(msg, trace) {{ | |
3865 | var msgStack = ['PHANTOM ERROR: ' + msg]; | |
3866 | if(trace && trace.length) {{ | |
3867 | msgStack.push('TRACE:'); | |
3868 | trace.forEach(function(t) {{ | |
3869 | msgStack.push(' -> ' + (t.file || t.sourceURL) + ': ' + t.line | |
3870 | + (t.function ? ' (in function ' + t.function +')' : '')); | |
3871 | }}); | |
3872 | }} | |
3873 | console.error(msgStack.join('\n')); | |
3874 | phantom.exit(1); | |
3875 | }}; | |
3876 | var page = require('webpage').create(); | |
3877 | var fs = require('fs'); | |
3878 | var read = {{ mode: 'r', charset: 'utf-8' }}; | |
3879 | var write = {{ mode: 'w', charset: 'utf-8' }}; | |
40e41780 TF |
3880 | JSON.parse(fs.read("{cookies}", read)).forEach(function(x) {{ |
3881 | phantom.addCookie(x); | |
3882 | }}); | |
da57ebaf TF |
3883 | page.settings.resourceTimeout = {timeout}; |
3884 | page.settings.userAgent = "{ua}"; | |
3885 | page.onLoadStarted = function() {{ | |
3886 | page.evaluate(function() {{ | |
3887 | delete window._phantom; | |
3888 | delete window.callPhantom; | |
3889 | }}); | |
3890 | }}; | |
3891 | var saveAndExit = function() {{ | |
3892 | fs.write("{html}", page.content, write); | |
40e41780 | 3893 | fs.write("{cookies}", JSON.stringify(phantom.cookies), write); |
da57ebaf TF |
3894 | phantom.exit(); |
3895 | }}; | |
3896 | page.onLoadFinished = function(status) {{ | |
3897 | if(page.url === "") {{ | |
3898 | page.setContent(fs.read("{html}", read), "{url}"); | |
3899 | }} | |
3900 | else {{ | |
3901 | {jscode} | |
3902 | }} | |
3903 | }}; | |
3904 | page.open(""); | |
3905 | ''' | |
3906 | ||
40e41780 | 3907 | _TMP_FILE_NAMES = ['script', 'html', 'cookies'] |
da57ebaf | 3908 | |
feee8d32 TF |
3909 | @staticmethod |
3910 | def _version(): | |
3911 | return get_exe_version('phantomjs', version_re=r'([0-9.]+)') | |
3912 | ||
98f9d873 | 3913 | def __init__(self, extractor, required_version=None, timeout=10000): |
da57ebaf TF |
3914 | self.exe = check_executable('phantomjs', ['-v']) |
3915 | if not self.exe: | |
3916 | raise ExtractorError('PhantomJS executable not found in PATH, ' | |
3917 | 'download it from http://phantomjs.org', | |
3918 | expected=True) | |
98f9d873 | 3919 | |
da57ebaf | 3920 | self.extractor = extractor |
98f9d873 TF |
3921 | |
3922 | if required_version: | |
feee8d32 | 3923 | version = self._version() |
98f9d873 TF |
3924 | if is_outdated_version(version, required_version): |
3925 | self.extractor._downloader.report_warning( | |
3926 | 'Your copy of PhantomJS is outdated, update it to version ' | |
3927 | '%s or newer if you encounter any errors.' % required_version) | |
3928 | ||
da57ebaf TF |
3929 | self.options = { |
3930 | 'timeout': timeout, | |
3931 | } | |
3932 | self._TMP_FILES = {} | |
3933 | for name in self._TMP_FILE_NAMES: | |
3934 | tmp = tempfile.NamedTemporaryFile(delete=False) | |
3935 | tmp.close() | |
3936 | self._TMP_FILES[name] = tmp | |
3937 | ||
3938 | def __del__(self): | |
3939 | for name in self._TMP_FILE_NAMES: | |
3940 | try: | |
3941 | os.remove(self._TMP_FILES[name].name) | |
3942 | except: | |
3943 | pass | |
3944 | ||
40e41780 TF |
3945 | def _save_cookies(self, url): |
3946 | cookies = cookie_jar_to_list(self.extractor._downloader.cookiejar) | |
3947 | for cookie in cookies: | |
3948 | if 'path' not in cookie: | |
3949 | cookie['path'] = '/' | |
3950 | if 'domain' not in cookie: | |
3951 | cookie['domain'] = compat_urlparse.urlparse(url).netloc | |
3952 | with open(self._TMP_FILES['cookies'].name, 'wb') as f: | |
3953 | f.write(json.dumps(cookies).encode('utf-8')) | |
3954 | ||
3955 | def _load_cookies(self): | |
3956 | with open(self._TMP_FILES['cookies'].name, 'rb') as f: | |
3957 | cookies = json.loads(f.read().decode('utf-8')) | |
3958 | for cookie in cookies: | |
3959 | if cookie['httponly'] is True: | |
4ed2d7b7 | 3960 | cookie['rest'] = {'httpOnly': None} |
40e41780 TF |
3961 | if 'expiry' in cookie: |
3962 | cookie['expire_time'] = cookie['expiry'] | |
3963 | self.extractor._set_cookie(**cookie) | |
3964 | ||
da57ebaf TF |
3965 | def get(self, url, html=None, video_id=None, note=None, note2='Executing JS on webpage', headers={}, jscode='saveAndExit();'): |
3966 | """ | |
3967 | Downloads webpage (if needed) and executes JS | |
4ed2d7b7 | 3968 | |
da57ebaf TF |
3969 | Params: |
3970 | url: website url | |
3971 | html: optional, html code of website | |
3972 | video_id: video id | |
3973 | note: optional, displayed when downloading webpage | |
3974 | note2: optional, displayed when executing JS | |
3975 | headers: custom http headers | |
3976 | jscode: code to be executed when page is loaded | |
4ed2d7b7 | 3977 | |
da57ebaf TF |
3978 | Returns tuple with: |
3979 | * downloaded website (after JS execution) | |
3980 | * anything you print with `console.log` (but not inside `page.execute`!) | |
4ed2d7b7 | 3981 | |
da57ebaf TF |
3982 | In most cases you don't need to add any `jscode`. |
3983 | It is executed in `page.onLoadFinished`. | |
3984 | `saveAndExit();` is mandatory, use it instead of `phantom.exit()` | |
3985 | It is possible to wait for some element on the webpage, for example: | |
3986 | var check = function() { | |
3987 | var elementFound = page.evaluate(function() { | |
3988 | return document.querySelector('#b.done') !== null; | |
3989 | }); | |
3990 | if(elementFound) | |
3991 | saveAndExit(); | |
3992 | else | |
3993 | window.setTimeout(check, 500); | |
3994 | } | |
4ed2d7b7 | 3995 | |
da57ebaf TF |
3996 | page.evaluate(function(){ |
3997 | document.querySelector('#a').click(); | |
3998 | }); | |
3999 | check(); | |
4000 | """ | |
4001 | if 'saveAndExit();' not in jscode: | |
4002 | raise ExtractorError('`saveAndExit();` not found in `jscode`') | |
4003 | if not html: | |
4004 | html = self.extractor._download_webpage(url, video_id, note=note, headers=headers) | |
4005 | with open(self._TMP_FILES['html'].name, 'wb') as f: | |
4006 | f.write(html.encode('utf-8')) | |
4007 | ||
40e41780 TF |
4008 | self._save_cookies(url) |
4009 | ||
da57ebaf TF |
4010 | replaces = self.options |
4011 | replaces['url'] = url | |
4012 | user_agent = headers.get('User-Agent') or std_headers['User-Agent'] | |
4013 | replaces['ua'] = user_agent.replace('"', '\\"') | |
4014 | replaces['jscode'] = jscode | |
4015 | ||
4016 | for x in self._TMP_FILE_NAMES: | |
4017 | replaces[x] = self._TMP_FILES[x].name.replace('\\', '\\\\').replace('"', '\\"') | |
4018 | ||
4019 | with open(self._TMP_FILES['script'].name, 'wb') as f: | |
4020 | f.write(self._TEMPLATE.format(**replaces).encode('utf-8')) | |
4021 | ||
4022 | if video_id is None: | |
4023 | self.extractor.to_screen('%s' % (note2,)) | |
4024 | else: | |
4025 | self.extractor.to_screen('%s: %s' % (video_id, note2)) | |
4026 | ||
4ed2d7b7 YCH |
4027 | p = subprocess.Popen([ |
4028 | self.exe, '--ssl-protocol=any', | |
4029 | self._TMP_FILES['script'].name | |
4030 | ], stdout=subprocess.PIPE, stderr=subprocess.PIPE) | |
da57ebaf TF |
4031 | out, err = p.communicate() |
4032 | if p.returncode != 0: | |
4ed2d7b7 YCH |
4033 | raise ExtractorError( |
4034 | 'Executing JS failed\n:' + encodeArgument(err)) | |
da57ebaf TF |
4035 | with open(self._TMP_FILES['html'].name, 'rb') as f: |
4036 | html = f.read().decode('utf-8') | |
40e41780 TF |
4037 | |
4038 | self._load_cookies() | |
4039 | ||
da57ebaf TF |
4040 | return (html, encodeArgument(out)) |
4041 | ||
c89267d3 | 4042 | |
0c265486 YCH |
4043 | def random_birthday(year_field, month_field, day_field): |
4044 | return { | |
4045 | year_field: str(random.randint(1950, 1995)), | |
4046 | month_field: str(random.randint(1, 12)), | |
4047 | day_field: str(random.randint(1, 31)), | |
4048 | } |